Oskar Douwe van der Wal commited on
Commit
7bc099d
·
1 Parent(s): 52534dd

New results

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. pythia-14m-seed0/step0/EleutherAI__pythia-14m/results_2024-08-15T12-48-54.868316.json +96 -0
  2. pythia-14m-seed0/step1/EleutherAI__pythia-14m/results_2024-08-15T12-50-21.093189.json +96 -0
  3. pythia-14m-seed0/step1000/EleutherAI__pythia-14m/results_2024-08-15T13-04-58.314501.json +96 -0
  4. pythia-14m-seed0/step10000/EleutherAI__pythia-14m/results_2024-08-15T13-18-05.667000.json +96 -0
  5. pythia-14m-seed0/step100000/EleutherAI__pythia-14m/results_2024-08-15T15-27-54.136872.json +96 -0
  6. pythia-14m-seed0/step101000/EleutherAI__pythia-14m/results_2024-08-15T15-29-20.379855.json +96 -0
  7. pythia-14m-seed0/step102000/EleutherAI__pythia-14m/results_2024-08-15T15-30-45.504373.json +96 -0
  8. pythia-14m-seed0/step103000/EleutherAI__pythia-14m/results_2024-08-15T15-32-11.742196.json +96 -0
  9. pythia-14m-seed0/step104000/EleutherAI__pythia-14m/results_2024-08-15T15-33-38.569825.json +96 -0
  10. pythia-14m-seed0/step105000/EleutherAI__pythia-14m/results_2024-08-15T15-35-05.211421.json +96 -0
  11. pythia-14m-seed0/step106000/EleutherAI__pythia-14m/results_2024-08-15T15-36-32.465444.json +96 -0
  12. pythia-14m-seed0/step107000/EleutherAI__pythia-14m/results_2024-08-15T15-38-00.033086.json +96 -0
  13. pythia-14m-seed0/step108000/EleutherAI__pythia-14m/results_2024-08-15T15-39-28.787439.json +96 -0
  14. pythia-14m-seed0/step109000/EleutherAI__pythia-14m/results_2024-08-15T15-40-55.857604.json +96 -0
  15. pythia-14m-seed0/step11000/EleutherAI__pythia-14m/results_2024-08-15T13-19-30.603325.json +96 -0
  16. pythia-14m-seed0/step110000/EleutherAI__pythia-14m/results_2024-08-15T15-42-21.675993.json +96 -0
  17. pythia-14m-seed0/step111000/EleutherAI__pythia-14m/results_2024-08-15T15-43-46.617900.json +96 -0
  18. pythia-14m-seed0/step112000/EleutherAI__pythia-14m/results_2024-08-15T15-45-12.514667.json +96 -0
  19. pythia-14m-seed0/step113000/EleutherAI__pythia-14m/results_2024-08-15T15-46-38.770492.json +96 -0
  20. pythia-14m-seed0/step114000/EleutherAI__pythia-14m/results_2024-08-15T15-48-05.260016.json +96 -0
  21. pythia-14m-seed0/step115000/EleutherAI__pythia-14m/results_2024-08-15T15-49-32.634482.json +96 -0
  22. pythia-14m-seed0/step116000/EleutherAI__pythia-14m/results_2024-08-15T15-51-01.124488.json +96 -0
  23. pythia-14m-seed0/step117000/EleutherAI__pythia-14m/results_2024-08-15T15-52-30.240629.json +96 -0
  24. pythia-14m-seed0/step118000/EleutherAI__pythia-14m/results_2024-08-15T15-53-56.953504.json +96 -0
  25. pythia-14m-seed0/step119000/EleutherAI__pythia-14m/results_2024-08-15T15-55-23.128925.json +96 -0
  26. pythia-14m-seed0/step12000/EleutherAI__pythia-14m/results_2024-08-15T13-20-56.302395.json +96 -0
  27. pythia-14m-seed0/step120000/EleutherAI__pythia-14m/results_2024-08-15T15-56-52.216443.json +96 -0
  28. pythia-14m-seed0/step121000/EleutherAI__pythia-14m/results_2024-08-15T15-58-17.456909.json +96 -0
  29. pythia-14m-seed0/step122000/EleutherAI__pythia-14m/results_2024-08-15T15-59-44.315496.json +96 -0
  30. pythia-14m-seed0/step123000/EleutherAI__pythia-14m/results_2024-08-15T16-01-10.335351.json +96 -0
  31. pythia-14m-seed0/step124000/EleutherAI__pythia-14m/results_2024-08-15T16-02-35.938574.json +96 -0
  32. pythia-14m-seed0/step125000/EleutherAI__pythia-14m/results_2024-08-15T16-04-02.754305.json +96 -0
  33. pythia-14m-seed0/step126000/EleutherAI__pythia-14m/results_2024-08-15T16-05-29.292044.json +96 -0
  34. pythia-14m-seed0/step127000/EleutherAI__pythia-14m/results_2024-08-15T16-06-56.433830.json +96 -0
  35. pythia-14m-seed0/step128/EleutherAI__pythia-14m/results_2024-08-15T13-00-34.735877.json +96 -0
  36. pythia-14m-seed0/step128000/EleutherAI__pythia-14m/results_2024-08-15T16-08-22.902858.json +96 -0
  37. pythia-14m-seed0/step129000/EleutherAI__pythia-14m/results_2024-08-15T16-09-48.457182.json +96 -0
  38. pythia-14m-seed0/step13000/EleutherAI__pythia-14m/results_2024-08-15T13-22-24.295538.json +96 -0
  39. pythia-14m-seed0/step130000/EleutherAI__pythia-14m/results_2024-08-15T16-11-14.614730.json +96 -0
  40. pythia-14m-seed0/step131000/EleutherAI__pythia-14m/results_2024-08-15T16-12-40.692551.json +96 -0
  41. pythia-14m-seed0/step132000/EleutherAI__pythia-14m/results_2024-08-15T16-14-06.187317.json +96 -0
  42. pythia-14m-seed0/step133000/EleutherAI__pythia-14m/results_2024-08-15T16-15-33.241417.json +96 -0
  43. pythia-14m-seed0/step134000/EleutherAI__pythia-14m/results_2024-08-15T16-17-00.743542.json +96 -0
  44. pythia-14m-seed0/step135000/EleutherAI__pythia-14m/results_2024-08-15T16-18-28.514020.json +96 -0
  45. pythia-14m-seed0/step136000/EleutherAI__pythia-14m/results_2024-08-15T16-19-54.399480.json +96 -0
  46. pythia-14m-seed0/step137000/EleutherAI__pythia-14m/results_2024-08-15T16-21-20.940162.json +96 -0
  47. pythia-14m-seed0/step138000/EleutherAI__pythia-14m/results_2024-08-15T16-22-46.483201.json +96 -0
  48. pythia-14m-seed0/step139000/EleutherAI__pythia-14m/results_2024-08-15T16-24-12.061095.json +96 -0
  49. pythia-14m-seed0/step14000/EleutherAI__pythia-14m/results_2024-08-15T13-23-49.471021.json +96 -0
  50. pythia-14m-seed0/step140000/EleutherAI__pythia-14m/results_2024-08-15T16-25-37.975454.json +96 -0
pythia-14m-seed0/step0/EleutherAI__pythia-14m/results_2024-08-15T12-48-54.868316.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 3507781.8105202965,
5
+ "perplexity_stderr,none": 338907.2426000304,
6
+ "acc,none": 0.0,
7
+ "acc_stderr,none": 0.0,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step0,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step0",
71
+ "model_sha": "bc249c50b4a381f99e9296f08a69b8562a31fde4",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723751264.2585666,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1133831.651542538,
94
+ "end_time": 1133915.107703607,
95
+ "total_evaluation_time_seconds": "83.45616106898524"
96
+ }
pythia-14m-seed0/step1/EleutherAI__pythia-14m/results_2024-08-15T12-50-21.093189.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 3507781.8105202965,
5
+ "perplexity_stderr,none": 338907.2426000304,
6
+ "acc,none": 0.0,
7
+ "acc_stderr,none": 0.0,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step1,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step1",
71
+ "model_sha": "71caa8178c55d927b2234a30bbabac6bd072cae8",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723751353.851652,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1133922.894150664,
94
+ "end_time": 1134001.333325669,
95
+ "total_evaluation_time_seconds": "78.43917500483803"
96
+ }
pythia-14m-seed0/step1000/EleutherAI__pythia-14m/results_2024-08-15T13-04-58.314501.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 195988.90526637834,
5
+ "perplexity_stderr,none": 12193.7268807034,
6
+ "acc,none": 0.0,
7
+ "acc_stderr,none": 0.0,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step1000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step1000",
71
+ "model_sha": "67ed4908dc8d91741d246c4d9fbfa03e78c8505d",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723752224.1879683,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1134793.204587141,
94
+ "end_time": 1134878.555076991,
95
+ "total_evaluation_time_seconds": "85.35048984992318"
96
+ }
pythia-14m-seed0/step10000/EleutherAI__pythia-14m/results_2024-08-15T13-18-05.667000.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1511.251720455259,
5
+ "perplexity_stderr,none": 70.29803394374686,
6
+ "acc,none": 0.1003299049097613,
7
+ "acc_stderr,none": 0.004185707878554855,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step10000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step10000",
71
+ "model_sha": "50c427ff77acdb2553ee434ffabe182f5be3c1e2",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723753014.309483,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1135582.228500084,
94
+ "end_time": 1135665.902803599,
95
+ "total_evaluation_time_seconds": "83.67430351488292"
96
+ }
pythia-14m-seed0/step100000/EleutherAI__pythia-14m/results_2024-08-15T15-27-54.136872.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 2480.572303985594,
5
+ "perplexity_stderr,none": 110.31519283833292,
6
+ "acc,none": 0.05142635358043858,
7
+ "acc_stderr,none": 0.0030770918924100804,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step100000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step100000",
71
+ "model_sha": "c5ab9636d47fcd482787a66ae2ef8a71002648b9",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723760806.4035344,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 2500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1143373.76838171,
94
+ "end_time": 1143454.366109349,
95
+ "total_evaluation_time_seconds": "80.59772763913497"
96
+ }
pythia-14m-seed0/step101000/EleutherAI__pythia-14m/results_2024-08-15T15-29-20.379855.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1936.7673989556679,
5
+ "perplexity_stderr,none": 89.14269698503178,
6
+ "acc,none": 0.07354938870560838,
7
+ "acc_stderr,none": 0.0036367440978540603,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step101000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step101000",
71
+ "model_sha": "2a340986858902da1c8563a6c273066aaccc8750",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723760893.3566766,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1143461.457184083,
94
+ "end_time": 1143540.619783163,
95
+ "total_evaluation_time_seconds": "79.16259908000939"
96
+ }
pythia-14m-seed0/step102000/EleutherAI__pythia-14m/results_2024-08-15T15-30-45.504373.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 2332.1576555643323,
5
+ "perplexity_stderr,none": 103.2466067379606,
6
+ "acc,none": 0.055695711236173104,
7
+ "acc_stderr,none": 0.0031950593588073433,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step102000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step102000",
71
+ "model_sha": "e1ef05f9ceddbf4c8b5eda7b6e1eb66da607dc76",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723760978.5723045,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1143547.70613283,
94
+ "end_time": 1143625.744282615,
95
+ "total_evaluation_time_seconds": "78.0381497850176"
96
+ }
pythia-14m-seed0/step103000/EleutherAI__pythia-14m/results_2024-08-15T15-32-11.742196.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 2799.559936979982,
5
+ "perplexity_stderr,none": 123.34505192306801,
6
+ "acc,none": 0.05491946438967592,
7
+ "acc_stderr,none": 0.0031740197788982893,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step103000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step103000",
71
+ "model_sha": "2f4b2967a8d878296de9228ba494cd18a59fe35c",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723761064.4161794,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1143633.446766386,
94
+ "end_time": 1143711.9823575,
95
+ "total_evaluation_time_seconds": "78.53559111384675"
96
+ }
pythia-14m-seed0/step104000/EleutherAI__pythia-14m/results_2024-08-15T15-33-38.569825.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 2003.2493640780494,
5
+ "perplexity_stderr,none": 89.98672601843188,
6
+ "acc,none": 0.06617504366388512,
7
+ "acc_stderr,none": 0.003463314484220094,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step104000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step104000",
71
+ "model_sha": "678191556681a68fbcd641ab2bd018f3634fd545",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723761150.9524848,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1143719.967749508,
94
+ "end_time": 1143798.810041786,
95
+ "total_evaluation_time_seconds": "78.8422922778409"
96
+ }
pythia-14m-seed0/step105000/EleutherAI__pythia-14m/results_2024-08-15T15-35-05.211421.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 2204.366599640772,
5
+ "perplexity_stderr,none": 101.36401689274162,
6
+ "acc,none": 0.06966815447312245,
7
+ "acc_stderr,none": 0.003546893672151897,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step105000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step105000",
71
+ "model_sha": "52a764e3d099f01bcf6079282458009fba8e9815",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723761237.5477817,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 2500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1143806.663444182,
94
+ "end_time": 1143885.451216583,
95
+ "total_evaluation_time_seconds": "78.78777240100317"
96
+ }
pythia-14m-seed0/step106000/EleutherAI__pythia-14m/results_2024-08-15T15-36-32.465444.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 2260.64934437262,
5
+ "perplexity_stderr,none": 101.92424646136189,
6
+ "acc,none": 0.06986221618474675,
7
+ "acc_stderr,none": 0.003551459735491157,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step106000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step106000",
71
+ "model_sha": "fb05d722aeeb0944d2f041429d68cd7a0f0aefb5",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723761324.5612192,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 2500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1143892.700193422,
94
+ "end_time": 1143972.705910608,
95
+ "total_evaluation_time_seconds": "80.00571718602441"
96
+ }
pythia-14m-seed0/step107000/EleutherAI__pythia-14m/results_2024-08-15T15-38-00.033086.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 2030.1912500934511,
5
+ "perplexity_stderr,none": 89.98267591591406,
6
+ "acc,none": 0.05918882204541044,
7
+ "acc_stderr,none": 0.003287631671733364,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step107000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step107000",
71
+ "model_sha": "e51d508ae0521ac7f4d1f9879ae7a45133c4c71f",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723761412.6764872,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 2500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1143979.738471963,
94
+ "end_time": 1144060.273198181,
95
+ "total_evaluation_time_seconds": "80.5347262180876"
96
+ }
pythia-14m-seed0/step108000/EleutherAI__pythia-14m/results_2024-08-15T15-39-28.787439.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1942.191674587457,
5
+ "perplexity_stderr,none": 88.17936208188293,
6
+ "acc,none": 0.07219095672423831,
7
+ "acc_stderr,none": 0.003605643396603442,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step108000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step108000",
71
+ "model_sha": "e6e41d34d4cea7f1b7b7a3c789f10c07e9ea998a",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723761500.0742729,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1144067.406098665,
94
+ "end_time": 1144149.027880051,
95
+ "total_evaluation_time_seconds": "81.62178138596937"
96
+ }
pythia-14m-seed0/step109000/EleutherAI__pythia-14m/results_2024-08-15T15-40-55.857604.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1856.8399370102884,
5
+ "perplexity_stderr,none": 85.6448562808026,
6
+ "acc,none": 0.07413157384048127,
7
+ "acc_stderr,none": 0.0036499618074347898,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step109000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step109000",
71
+ "model_sha": "1406e20c840393345a50b6b8eaa261e2a8afa13f",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723761588.1894023,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1144156.112510001,
94
+ "end_time": 1144236.097111593,
95
+ "total_evaluation_time_seconds": "79.98460159194656"
96
+ }
pythia-14m-seed0/step11000/EleutherAI__pythia-14m/results_2024-08-15T13-19-30.603325.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1336.2520161255652,
5
+ "perplexity_stderr,none": 60.98331616528402,
6
+ "acc,none": 0.10285270716087716,
7
+ "acc_stderr,none": 0.004232059957170525,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step11000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step11000",
71
+ "model_sha": "23b9f9586a7ce1aa3b85566361c5471f05e36f30",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723753103.9197896,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1135673.004151162,
94
+ "end_time": 1135750.843636572,
95
+ "total_evaluation_time_seconds": "77.83948540990241"
96
+ }
pythia-14m-seed0/step110000/EleutherAI__pythia-14m/results_2024-08-15T15-42-21.675993.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1655.1447995081642,
5
+ "perplexity_stderr,none": 76.78861282594242,
6
+ "acc,none": 0.08092373374733165,
7
+ "acc_stderr,none": 0.0037994943646444323,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step110000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step110000",
71
+ "model_sha": "4f102eb31047258ae88f77a33ad7a1d22107ad7f",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723761674.655817,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1144243.183542134,
94
+ "end_time": 1144321.915977644,
95
+ "total_evaluation_time_seconds": "78.73243551002815"
96
+ }
pythia-14m-seed0/step111000/EleutherAI__pythia-14m/results_2024-08-15T15-43-46.617900.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1577.7555976392246,
5
+ "perplexity_stderr,none": 73.319450716533,
6
+ "acc,none": 0.08286435086357462,
7
+ "acc_stderr,none": 0.0038407207037192497,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step111000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step111000",
71
+ "model_sha": "99137989374bfe1ecb1a5e64b3d80ce774eec113",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723761759.958622,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 2500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1144328.95044279,
94
+ "end_time": 1144406.857867798,
95
+ "total_evaluation_time_seconds": "77.90742500801571"
96
+ }
pythia-14m-seed0/step112000/EleutherAI__pythia-14m/results_2024-08-15T15-45-12.514667.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1759.9239677979742,
5
+ "perplexity_stderr,none": 81.70703449195138,
6
+ "acc,none": 0.07490782068697846,
7
+ "acc_stderr,none": 0.0036674834158285856,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step112000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step112000",
71
+ "model_sha": "9774065c9f5a45ad7788ad185541011cbb0862d9",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723761845.0600786,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1144414.160817905,
94
+ "end_time": 1144492.75489787,
95
+ "total_evaluation_time_seconds": "78.59407996479422"
96
+ }
pythia-14m-seed0/step113000/EleutherAI__pythia-14m/results_2024-08-15T15-46-38.770492.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1588.236297320137,
5
+ "perplexity_stderr,none": 72.93322208929094,
6
+ "acc,none": 0.07490782068697846,
7
+ "acc_stderr,none": 0.0036674834158285856,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step113000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step113000",
71
+ "model_sha": "550e364271f369edb9d4a144425830fdedee75cd",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723761930.9352212,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1144499.950961815,
94
+ "end_time": 1144579.010207957,
95
+ "total_evaluation_time_seconds": "79.05924614192918"
96
+ }
pythia-14m-seed0/step114000/EleutherAI__pythia-14m/results_2024-08-15T15-48-05.260016.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1702.065327436129,
5
+ "perplexity_stderr,none": 78.56178715398188,
6
+ "acc,none": 0.07568406753347565,
7
+ "acc_stderr,none": 0.0036848899715327707,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step114000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step114000",
71
+ "model_sha": "9c800cd9948d544aff8957bed479581938a7d59c",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723762017.8288126,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1144586.610589841,
94
+ "end_time": 1144665.499988281,
95
+ "total_evaluation_time_seconds": "78.88939844002016"
96
+ }
pythia-14m-seed0/step115000/EleutherAI__pythia-14m/results_2024-08-15T15-49-32.634482.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1653.5844188453527,
5
+ "perplexity_stderr,none": 76.53860729699839,
6
+ "acc,none": 0.07471375897535416,
7
+ "acc_stderr,none": 0.003663113864225891,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step115000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step115000",
71
+ "model_sha": "84565561177046e994cc5d236a3f526cb0a1494e",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723762104.9888885,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1144673.793967198,
94
+ "end_time": 1144752.874586123,
95
+ "total_evaluation_time_seconds": "79.08061892492697"
96
+ }
pythia-14m-seed0/step116000/EleutherAI__pythia-14m/results_2024-08-15T15-51-01.124488.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1670.6856831411806,
5
+ "perplexity_stderr,none": 76.90491802120467,
6
+ "acc,none": 0.07432563555210557,
7
+ "acc_stderr,none": 0.0036543530860415506,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step116000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step116000",
71
+ "model_sha": "90fd8b63997a429c96f4d13a9d5290cb9621af6e",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723762193.411083,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1144760.733506932,
94
+ "end_time": 1144841.363911434,
95
+ "total_evaluation_time_seconds": "80.6304045021534"
96
+ }
pythia-14m-seed0/step117000/EleutherAI__pythia-14m/results_2024-08-15T15-52-30.240629.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1754.3725242092196,
5
+ "perplexity_stderr,none": 81.01434898802123,
6
+ "acc,none": 0.08072967203570736,
7
+ "acc_stderr,none": 0.003795336508106496,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step117000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step117000",
71
+ "model_sha": "b205c539eccb75ec24cffe6c918b8eb5419e9e1f",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723762281.6944668,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1144849.179324036,
94
+ "end_time": 1144930.480355115,
95
+ "total_evaluation_time_seconds": "81.30103107891046"
96
+ }
pythia-14m-seed0/step118000/EleutherAI__pythia-14m/results_2024-08-15T15-53-56.953504.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1758.961983552553,
5
+ "perplexity_stderr,none": 80.65986006671504,
6
+ "acc,none": 0.07587812924509994,
7
+ "acc_stderr,none": 0.0036892238245559935,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step118000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step118000",
71
+ "model_sha": "0cd8490766c1ec80c1f9aa970de646bd573192d1",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723762369.5567913,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1144937.75132631,
94
+ "end_time": 1145017.193742402,
95
+ "total_evaluation_time_seconds": "79.44241609191522"
96
+ }
pythia-14m-seed0/step119000/EleutherAI__pythia-14m/results_2024-08-15T15-55-23.128925.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1654.0146622462084,
5
+ "perplexity_stderr,none": 76.2436881269689,
6
+ "acc,none": 0.07665437609159713,
7
+ "acc_stderr,none": 0.0037064888417067503,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step119000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step119000",
71
+ "model_sha": "06f016ef45fd4df8e4118bec1aaaa17ce539d8eb",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723762455.409844,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1145024.27587102,
94
+ "end_time": 1145103.369137769,
95
+ "total_evaluation_time_seconds": "79.09326674905606"
96
+ }
pythia-14m-seed0/step12000/EleutherAI__pythia-14m/results_2024-08-15T13-20-56.302395.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1345.9520903737475,
5
+ "perplexity_stderr,none": 62.175751078363405,
6
+ "acc,none": 0.09179118959829226,
7
+ "acc_stderr,none": 0.004022586828929574,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step12000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step12000",
71
+ "model_sha": "677c73e05b0353916ad35c85c4f4609dce0348af",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723753189.099923,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 2200.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1135757.985463659,
94
+ "end_time": 1135836.540199971,
95
+ "total_evaluation_time_seconds": "78.55473631201312"
96
+ }
pythia-14m-seed0/step120000/EleutherAI__pythia-14m/results_2024-08-15T15-56-52.216443.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1885.6556307949313,
5
+ "perplexity_stderr,none": 84.00187419647823,
6
+ "acc,none": 0.06462254997089074,
7
+ "acc_stderr,none": 0.003425291731867342,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step120000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step120000",
71
+ "model_sha": "379be14e11d7e3ef68044bbbcc8890e192c013b9",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723762541.51547,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1145110.479581879,
94
+ "end_time": 1145192.456666261,
95
+ "total_evaluation_time_seconds": "81.97708438197151"
96
+ }
pythia-14m-seed0/step121000/EleutherAI__pythia-14m/results_2024-08-15T15-58-17.456909.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1760.6364312354708,
5
+ "perplexity_stderr,none": 81.10863398516484,
6
+ "acc,none": 0.07781874636134291,
7
+ "acc_stderr,none": 0.0037321778637123253,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step121000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step121000",
71
+ "model_sha": "b4462c932e24f3271ec4e0819a6af0af31105a9a",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723762630.560148,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1145199.569018539,
94
+ "end_time": 1145277.697085722,
95
+ "total_evaluation_time_seconds": "78.12806718284264"
96
+ }
pythia-14m-seed0/step122000/EleutherAI__pythia-14m/results_2024-08-15T15-59-44.315496.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1614.3906537485614,
5
+ "perplexity_stderr,none": 75.54730571478007,
6
+ "acc,none": 0.0836405977100718,
7
+ "acc_stderr,none": 0.0038570347957365973,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step122000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step122000",
71
+ "model_sha": "d05207ca4e2e9503facd5d5f03130fec1accc174",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723762715.8275197,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1145284.959185133,
94
+ "end_time": 1145364.550398592,
95
+ "total_evaluation_time_seconds": "79.59121345891617"
96
+ }
pythia-14m-seed0/step123000/EleutherAI__pythia-14m/results_2024-08-15T16-01-10.335351.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 2038.1899485513527,
5
+ "perplexity_stderr,none": 91.65050326709839,
6
+ "acc,none": 0.066757228798758,
7
+ "acc_stderr,none": 0.0034774311506045566,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step123000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step123000",
71
+ "model_sha": "bf4ec8affb1219b69a78d3ea02f4300e65c2627e",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723762802.8322046,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1145371.838932145,
94
+ "end_time": 1145450.57434153,
95
+ "total_evaluation_time_seconds": "78.73540938505903"
96
+ }
pythia-14m-seed0/step124000/EleutherAI__pythia-14m/results_2024-08-15T16-02-35.938574.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1899.7256494472654,
5
+ "perplexity_stderr,none": 85.27743433714627,
6
+ "acc,none": 0.0671453522220066,
7
+ "acc_stderr,none": 0.00348680002818163,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step124000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step124000",
71
+ "model_sha": "e6fb05f7d7a0437fe0bed7c289eeec6034f3ec93",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723762888.69861,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1145457.648579555,
94
+ "end_time": 1145536.161768229,
95
+ "total_evaluation_time_seconds": "78.51318867388181"
96
+ }
pythia-14m-seed0/step125000/EleutherAI__pythia-14m/results_2024-08-15T16-04-02.754305.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1716.8749780802998,
5
+ "perplexity_stderr,none": 79.70990558710086,
6
+ "acc,none": 0.07607219095672424,
7
+ "acc_stderr,none": 0.0036935506133658606,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step125000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step125000",
71
+ "model_sha": "460b51a3b8394f0bbcc5f839d989f2e19a262a9f",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723762975.4865017,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 2500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1145543.265744231,
94
+ "end_time": 1145622.994148452,
95
+ "total_evaluation_time_seconds": "79.72840422112495"
96
+ }
pythia-14m-seed0/step126000/EleutherAI__pythia-14m/results_2024-08-15T16-05-29.292044.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1648.377668467864,
5
+ "perplexity_stderr,none": 76.19708183106972,
6
+ "acc,none": 0.07704249951484572,
7
+ "acc_stderr,none": 0.003715079456679115,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step126000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step126000",
71
+ "model_sha": "7bdc419ff72691c556674155c1a83bc6e88cd742",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723763062.0582552,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 2500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1145630.007922808,
94
+ "end_time": 1145709.532083368,
95
+ "total_evaluation_time_seconds": "79.52416055998765"
96
+ }
pythia-14m-seed0/step127000/EleutherAI__pythia-14m/results_2024-08-15T16-06-56.433830.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1746.9065161719782,
5
+ "perplexity_stderr,none": 80.67923580716102,
6
+ "acc,none": 0.07898311663108869,
7
+ "acc_stderr,none": 0.003757621238955833,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step127000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step127000",
71
+ "model_sha": "0ebce7cc0fad8445578b776dae9f0c2dd9017d50",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723763149.0984063,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1145716.698455785,
94
+ "end_time": 1145796.673910495,
95
+ "total_evaluation_time_seconds": "79.97545470995829"
96
+ }
pythia-14m-seed0/step128/EleutherAI__pythia-14m/results_2024-08-15T13-00-34.735877.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 2154450.549794474,
5
+ "perplexity_stderr,none": 190529.88059195384,
6
+ "acc,none": 0.0,
7
+ "acc_stderr,none": 0.0,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step128,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step128",
71
+ "model_sha": "712cec71d5646cb88f76ee92029c12132ccd754c",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723751967.8752835,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1134536.513561532,
94
+ "end_time": 1134614.975972693,
95
+ "total_evaluation_time_seconds": "78.46241116104648"
96
+ }
pythia-14m-seed0/step128000/EleutherAI__pythia-14m/results_2024-08-15T16-08-22.902858.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1695.180554031583,
5
+ "perplexity_stderr,none": 77.5396989490735,
6
+ "acc,none": 0.06908596933824956,
7
+ "acc_stderr,none": 0.003533147663272781,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step128000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step128000",
71
+ "model_sha": "698a9b44b5707c15193cb9462f02c1952bf7ef29",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723763234.9688888,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1145803.817017457,
94
+ "end_time": 1145883.143051203,
95
+ "total_evaluation_time_seconds": "79.32603374612518"
96
+ }
pythia-14m-seed0/step129000/EleutherAI__pythia-14m/results_2024-08-15T16-09-48.457182.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1748.2060562452823,
5
+ "perplexity_stderr,none": 80.70472934165429,
6
+ "acc,none": 0.06908596933824956,
7
+ "acc_stderr,none": 0.003533147663272781,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step129000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step129000",
71
+ "model_sha": "9e3f931ba445a319b202efae7ca3b6624b085ae4",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723763321.257125,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1145890.282894991,
94
+ "end_time": 1145968.695761718,
95
+ "total_evaluation_time_seconds": "78.41286672698334"
96
+ }
pythia-14m-seed0/step13000/EleutherAI__pythia-14m/results_2024-08-15T13-22-24.295538.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1245.8883648802528,
5
+ "perplexity_stderr,none": 57.69527819537371,
6
+ "acc,none": 0.10576363283524161,
7
+ "acc_stderr,none": 0.004284561852149411,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step13000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step13000",
71
+ "model_sha": "51725d80c402ddd0b1ef23cb322f0fff2b7ab057",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723753276.8532705,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 2500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1135845.439155096,
94
+ "end_time": 1135924.535229917,
95
+ "total_evaluation_time_seconds": "79.0960748209618"
96
+ }
pythia-14m-seed0/step130000/EleutherAI__pythia-14m/results_2024-08-15T16-11-14.614730.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1615.9022166310167,
5
+ "perplexity_stderr,none": 74.77044537396877,
6
+ "acc,none": 0.08111779545895595,
7
+ "acc_stderr,none": 0.0038036457543544435,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step130000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step130000",
71
+ "model_sha": "c5a452c0e21b2beb5583459f763449ed56caa5c2",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723763407.04308,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1145975.991028769,
94
+ "end_time": 1146054.855413207,
95
+ "total_evaluation_time_seconds": "78.86438443814404"
96
+ }
pythia-14m-seed0/step131000/EleutherAI__pythia-14m/results_2024-08-15T16-12-40.692551.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1429.6528872004685,
5
+ "perplexity_stderr,none": 66.43871183549871,
6
+ "acc,none": 0.08888026392392781,
7
+ "acc_stderr,none": 0.003964628217610132,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step131000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step131000",
71
+ "model_sha": "4facff6a23be64c8b725719006ddd396b322144f",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723763493.1769783,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1146062.15420514,
94
+ "end_time": 1146140.932159504,
95
+ "total_evaluation_time_seconds": "78.77795436396264"
96
+ }
pythia-14m-seed0/step132000/EleutherAI__pythia-14m/results_2024-08-15T16-14-06.187317.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1744.4924353855185,
5
+ "perplexity_stderr,none": 81.16053763865396,
6
+ "acc,none": 0.07180283330098972,
7
+ "acc_stderr,none": 0.0035966897962539787,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step132000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step132000",
71
+ "model_sha": "d32e674e5bcaaabda91936dc3c8d6ebc2873fa18",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723763579.0232463,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1146148.026271967,
94
+ "end_time": 1146226.427460756,
95
+ "total_evaluation_time_seconds": "78.40118878893554"
96
+ }
pythia-14m-seed0/step133000/EleutherAI__pythia-14m/results_2024-08-15T16-15-33.241417.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1654.4108715342993,
5
+ "perplexity_stderr,none": 77.65858495677924,
6
+ "acc,none": 0.08228216572870173,
7
+ "acc_stderr,none": 0.0038284194650137684,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step133000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step133000",
71
+ "model_sha": "c0fe6645a6d36cae5975f2d5ba418a058ab1023e",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723763665.9182217,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 2500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1146233.538208775,
94
+ "end_time": 1146313.481529114,
95
+ "total_evaluation_time_seconds": "79.94332033907995"
96
+ }
pythia-14m-seed0/step134000/EleutherAI__pythia-14m/results_2024-08-15T16-17-00.743542.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1714.602030249706,
5
+ "perplexity_stderr,none": 78.98081491097976,
6
+ "acc,none": 0.07762468464971861,
7
+ "acc_stderr,none": 0.003727913555390672,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step134000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step134000",
71
+ "model_sha": "6aacc13c38d2aa88207071a8868e41c6e22cc82b",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723763753.4132254,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1146320.668391112,
94
+ "end_time": 1146400.983705253,
95
+ "total_evaluation_time_seconds": "80.31531414110214"
96
+ }
pythia-14m-seed0/step135000/EleutherAI__pythia-14m/results_2024-08-15T16-18-28.514020.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1645.9272622179124,
5
+ "perplexity_stderr,none": 76.43349328751616,
6
+ "acc,none": 0.07743062293809432,
7
+ "acc_stderr,none": 0.0037236424005269012,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step135000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step135000",
71
+ "model_sha": "95509b738e052a84f0d6741472fbec47268ada6d",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723763840.8905025,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1146408.240149021,
94
+ "end_time": 1146488.754164905,
95
+ "total_evaluation_time_seconds": "80.51401588413864"
96
+ }
pythia-14m-seed0/step136000/EleutherAI__pythia-14m/results_2024-08-15T16-19-54.399480.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1960.7084303102972,
5
+ "perplexity_stderr,none": 88.24537878239123,
6
+ "acc,none": 0.06442848825926645,
7
+ "acc_stderr,none": 0.003420499552612167,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step136000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step136000",
71
+ "model_sha": "2dd2b497d1fa1285e9288538bf03af70ea88c3f1",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723763927.2161357,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1146495.818538224,
94
+ "end_time": 1146574.639468531,
95
+ "total_evaluation_time_seconds": "78.82093030679971"
96
+ }
pythia-14m-seed0/step137000/EleutherAI__pythia-14m/results_2024-08-15T16-21-20.940162.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1592.062887278126,
5
+ "perplexity_stderr,none": 73.0489453312968,
6
+ "acc,none": 0.07762468464971861,
7
+ "acc_stderr,none": 0.003727913555390672,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step137000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step137000",
71
+ "model_sha": "f9cd17305d5cfb4f3776c1ea96d9b0f00fe32ede",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723764012.7023847,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1146581.770022683,
94
+ "end_time": 1146661.180396564,
95
+ "total_evaluation_time_seconds": "79.41037388099357"
96
+ }
pythia-14m-seed0/step138000/EleutherAI__pythia-14m/results_2024-08-15T16-22-46.483201.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1616.2517548162928,
5
+ "perplexity_stderr,none": 74.09170027590162,
6
+ "acc,none": 0.07490782068697846,
7
+ "acc_stderr,none": 0.0036674834158285856,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step138000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step138000",
71
+ "model_sha": "720f854dbe30d974b7b495c0b076e452aab854b9",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723764099.134047,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1146668.278176658,
94
+ "end_time": 1146746.723473584,
95
+ "total_evaluation_time_seconds": "78.44529692595825"
96
+ }
pythia-14m-seed0/step139000/EleutherAI__pythia-14m/results_2024-08-15T16-24-12.061095.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1698.8500597374266,
5
+ "perplexity_stderr,none": 78.62022294671738,
6
+ "acc,none": 0.07549000582185135,
7
+ "acc_stderr,none": 0.00368054902934182,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step139000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step139000",
71
+ "model_sha": "7dde889802ebc8e9b3498bcc7d02ce199d9d4d76",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723764184.9015696,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1146753.933885823,
94
+ "end_time": 1146832.300943617,
95
+ "total_evaluation_time_seconds": "78.36705779400654"
96
+ }
pythia-14m-seed0/step14000/EleutherAI__pythia-14m/results_2024-08-15T13-23-49.471021.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1334.3251049658206,
5
+ "perplexity_stderr,none": 62.74215047810956,
6
+ "acc,none": 0.10304676887250146,
7
+ "acc_stderr,none": 0.004235592410767947,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step14000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step14000",
71
+ "model_sha": "d1cd73f742163cf3d733776a53422349d945caa8",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723753362.6691036,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 2500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1135931.618820009,
94
+ "end_time": 1136009.711027219,
95
+ "total_evaluation_time_seconds": "78.09220720990561"
96
+ }
pythia-14m-seed0/step140000/EleutherAI__pythia-14m/results_2024-08-15T16-25-37.975454.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "lambada_openai": {
4
+ "perplexity,none": 1635.2112969586547,
5
+ "perplexity_stderr,none": 75.2598585640202,
6
+ "acc,none": 0.07490782068697846,
7
+ "acc_stderr,none": 0.0036674834158285856,
8
+ "alias": "lambada_openai"
9
+ }
10
+ },
11
+ "group_subtasks": {
12
+ "lambada_openai": []
13
+ },
14
+ "configs": {
15
+ "lambada_openai": {
16
+ "task": "lambada_openai",
17
+ "group": [
18
+ "lambada"
19
+ ],
20
+ "dataset_path": "EleutherAI/lambada_openai",
21
+ "dataset_name": "default",
22
+ "dataset_kwargs": {
23
+ "trust_remote_code": true
24
+ },
25
+ "test_split": "test",
26
+ "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
27
+ "doc_to_target": "{{' '+text.split(' ')[-1]}}",
28
+ "description": "",
29
+ "target_delimiter": " ",
30
+ "fewshot_delimiter": "\n\n",
31
+ "num_fewshot": 0,
32
+ "metric_list": [
33
+ {
34
+ "metric": "perplexity",
35
+ "aggregation": "perplexity",
36
+ "higher_is_better": false
37
+ },
38
+ {
39
+ "metric": "acc",
40
+ "aggregation": "mean",
41
+ "higher_is_better": true
42
+ }
43
+ ],
44
+ "output_type": "loglikelihood",
45
+ "repeats": 1,
46
+ "should_decontaminate": true,
47
+ "doc_to_decontamination_query": "{{text}}",
48
+ "metadata": {
49
+ "version": 1.0
50
+ }
51
+ }
52
+ },
53
+ "versions": {
54
+ "lambada_openai": 1.0
55
+ },
56
+ "n-shot": {
57
+ "lambada_openai": 0
58
+ },
59
+ "n-samples": {
60
+ "lambada_openai": {
61
+ "original": 5153,
62
+ "effective": 5153
63
+ }
64
+ },
65
+ "config": {
66
+ "model": "hf",
67
+ "model_args": "pretrained=EleutherAI/pythia-14m,revision=step140000,",
68
+ "model_num_parameters": 14067712,
69
+ "model_dtype": "torch.float16",
70
+ "model_revision": "step140000",
71
+ "model_sha": "5bb0fea58b8b2712c579d791ff1c8bf2790cca10",
72
+ "batch_size": "128",
73
+ "batch_sizes": [],
74
+ "device": "cuda",
75
+ "use_cache": null,
76
+ "limit": null,
77
+ "bootstrap_iters": 100000,
78
+ "gen_kwargs": null,
79
+ "random_seed": 0,
80
+ "numpy_seed": 1234,
81
+ "torch_seed": 1234,
82
+ "fewshot_seed": 1234
83
+ },
84
+ "git_hash": "51a7ca9",
85
+ "date": 1723764270.5154667,
86
+ "pretty_env_info": "PyTorch version: 2.3.0+cu121\nIs debug build: False\nCUDA used to build PyTorch: 12.1\nROCM used to build PyTorch: N/A\n\nOS: CentOS Linux release 7.9.2009 (Core) (x86_64)\nGCC version: (GCC) 12.1.0\nClang version: Could not collect\nCMake version: Could not collect\nLibc version: glibc-2.17\n\nPython version: 3.12.1 (main, Jan 12 2024, 16:49:08) [GCC 12.1.0] (64-bit runtime)\nPython platform: Linux-3.10.0-1160.119.1.el7.tuxcare.els2.x86_64-x86_64-with-glibc2.17\nIs CUDA available: True\nCUDA runtime version: 12.4.99\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: GPU 0: Tesla V100S-PCIE-32GB\nNvidia driver version: 550.90.07\ncuDNN version: Could not collect\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nByte Order: Little Endian\nCPU(s): 32\nOn-line CPU(s) list: 0-31\nThread(s) per core: 1\nCore(s) per socket: 32\nSocket(s): 1\nNUMA node(s): 2\nVendor ID: AuthenticAMD\nCPU family: 23\nModel: 49\nModel name: AMD EPYC 7502P 32-Core Processor\nStepping: 0\nCPU MHz: 1500.000\nCPU max MHz: 2500.0000\nCPU min MHz: 1500.0000\nBogoMIPS: 5000.08\nVirtualization: AMD-V\nL1d cache: 32K\nL1i cache: 32K\nL2 cache: 512K\nL3 cache: 16384K\nNUMA node0 CPU(s): 0-15\nNUMA node1 CPU(s): 16-31\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc art rep_good nopl nonstop_tsc extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb cat_l3 cdp_l3 hw_pstate sme ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif umip overflow_recov succor smca\n\nVersions of relevant libraries:\n[pip3] numpy==1.26.4\n[pip3] torch==2.3.0\n[conda] Could not collect",
87
+ "transformers_version": "4.40.2",
88
+ "upper_git_hash": null,
89
+ "task_hashes": {},
90
+ "model_source": "hf",
91
+ "model_name": "EleutherAI/pythia-14m",
92
+ "model_name_sanitized": "EleutherAI__pythia-14m",
93
+ "start_time": 1146839.438632164,
94
+ "end_time": 1146918.215870824,
95
+ "total_evaluation_time_seconds": "78.77723866002634"
96
+ }