Hugging Face's logo

LeoCordoba
/
mt5-small-mlsum

Summarization
Transformers
PyTorch
JAX
Safetensors
Spanish
mt5
text2text-generation
sagemaker
spanish
Eval Results
Inference Endpoints
Model card Files Community
1
LeoCordoba commited on
Commit
d628a5c
•
1 Parent(s): 647069e

commit files to HF hub

Browse files
Files changed (14) hide show
  1. README.md +97 -0
  2. all_results.json +40 -0
  3. config.json +27 -0
  4. eval_results.json +16 -0
  5. metadata.json +1 -0
  6. pytorch_model.bin +3 -0
  7. special_tokens_map.json +1 -0
  8. spiece.model +3 -0
  9. test_generations.txt +0 -0
  10. test_results.json +15 -0
  11. tokenizer_config.json +1 -0
  12. train_results.json +14 -0
  13. trainer_state.json +719 -0
  14. training_args.bin +3 -0
README.md ADDED
@@ -0,0 +1,97 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+ ---
3
+ language: en
4
+ tags:
5
+ - sagemaker
6
+ - mt5
7
+ - summarization
8
+ - spanish
9
+ license: apache-2.0
10
+ datasets:
11
+ - mlsum - es
12
+ model-index:
13
+ - name: `mt5-small-mlsum`
14
+ results:
15
+ - task:
16
+ name: Abstractive Text Summarization
17
+ type: abstractive-text-summarization
18
+ dataset:
19
+ name: "MLSUM: MultiLingual SUMmarization dataset (Spanish)"
20
+ type: mlsum
21
+ metrics:
22
+ - name: Validation ROGUE-1
23
+ type: rogue-1
24
+ value: 26.4352
25
+ - name: Validation ROGUE-2
26
+ type: rogue-2
27
+ value: 8.9293
28
+ - name: Validation ROGUE-L
29
+ type: rogue-l
30
+ value: 21.2622
31
+ - name: Validation ROGUE-LSUM
32
+ type: rogue-lsum
33
+ value: 21.5518
34
+ - name: Test ROGUE-1
35
+ type: rogue-1
36
+ value: 26.0756
37
+ - name: Test ROGUE-2
38
+ type: rogue-2
39
+ value: 8.4669
40
+ - name: Test ROGUE-L
41
+ type: rogue-l
42
+ value: 20.8167
43
+ - name: Validation ROGUE-LSUM
44
+ type: rogue-lsum
45
+ value: 21.0822
46
+ widget:
47
+ - text: |
48
+ Jeff: Can I train a 🤗 Transformers model on Amazon SageMaker?
49
+ Philipp: Sure you can use the new Hugging Face Deep Learning Container.
50
+ Jeff: ok.
51
+ Jeff: and how can I get started?
52
+ Jeff: where can I find documentation?
53
+ Philipp: ok, ok you can find everything here. https://huggingface.co/blog/the-partnership-amazon-sagemaker-and-hugging-face
54
+ ---
55
+ ## `mt5-small-mlsum`
56
+ This model was trained using Amazon SageMaker and the new Hugging Face Deep Learning container.
57
+ For more information look at:
58
+ - [🤗 Transformers Documentation: Amazon SageMaker](https://huggingface.co/transformers/sagemaker.html)
59
+ - [Example Notebooks](https://github.com/huggingface/notebooks/tree/master/sagemaker)
60
+ - [Amazon SageMaker documentation for Hugging Face](https://docs.aws.amazon.com/sagemaker/latest/dg/hugging-face.html)
61
+ - [Python SDK SageMaker documentation for Hugging Face](https://sagemaker.readthedocs.io/en/stable/frameworks/huggingface/index.html)
62
+ - [Deep Learning Container](https://github.com/aws/deep-learning-containers/blob/master/available_images.md#huggingface-training-containers)
63
+ ## Hyperparameters
64
+ {
65
+ "dataset_config": "es",
66
+ "dataset_name": "mlsum",
67
+ "do_eval": true,
68
+ "do_predict": true,
69
+ "do_train": true,
70
+ "fp16": true,
71
+ "max_target_length": 64,
72
+ "model_name_or_path": "google/mt5-small",
73
+ "num_train_epochs": 10,
74
+ "output_dir": "/opt/ml/checkpoints",
75
+ "per_device_eval_batch_size": 4,
76
+ "per_device_train_batch_size": 4,
77
+ "predict_with_generate": true,
78
+ "sagemaker_container_log_level": 20,
79
+ "sagemaker_program": "run_summarization.py",
80
+ "save_strategy": "epoch",
81
+ "seed": 7,
82
+ "summary_column": "summary",
83
+ "text_column": "text"
84
+ }
85
+ ## Usage
86
+
87
+ ## Results
88
+ | key | value |
89
+ | --- | ----- |
90
+ | eval_rouge1 | 26.4352 |
91
+ | eval_rouge2 | 8.9293 |
92
+ | eval_rougeL | 21.2622 |
93
+ | eval_rougeLsum | 21.5518 |
94
+ | test_rouge1 | 26.0756 |
95
+ | test_rouge2 | 8.4669 |
96
+ | test_rougeL | 20.8167 |
97
+ | test_rougeLsum | 21.0822 |
all_results.json ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 7.0,
3
+ "eval_gen_len": 33.9532,
4
+ "eval_loss": 1.9251619577407837,
5
+ "eval_mem_cpu_alloc_delta": 2314445767,
6
+ "eval_mem_cpu_peaked_delta": 26998734,
7
+ "eval_mem_gpu_alloc_delta": 0,
8
+ "eval_mem_gpu_peaked_delta": 3355142144,
9
+ "eval_rouge1": 26.4352,
10
+ "eval_rouge2": 8.9293,
11
+ "eval_rougeL": 21.2622,
12
+ "eval_rougeLsum": 21.5518,
13
+ "eval_runtime": 1054.0821,
14
+ "eval_samples": 10358,
15
+ "eval_samples_per_second": 9.827,
16
+ "init_mem_cpu_alloc_delta": 177892,
17
+ "init_mem_cpu_peaked_delta": 18282,
18
+ "init_mem_gpu_alloc_delta": 1200707584,
19
+ "init_mem_gpu_peaked_delta": 0,
20
+ "test_gen_len": 33.803,
21
+ "test_loss": 2.005567789077759,
22
+ "test_mem_cpu_alloc_delta": 3104676530,
23
+ "test_mem_cpu_peaked_delta": 36246285,
24
+ "test_mem_gpu_alloc_delta": 0,
25
+ "test_mem_gpu_peaked_delta": 3357975040,
26
+ "test_rouge1": 26.0756,
27
+ "test_rouge2": 8.4669,
28
+ "test_rougeL": 20.8167,
29
+ "test_rougeLsum": 21.0822,
30
+ "test_runtime": 1405.8563,
31
+ "test_samples": 13920,
32
+ "test_samples_per_second": 9.901,
33
+ "train_mem_cpu_alloc_delta": 9871729,
34
+ "train_mem_cpu_peaked_delta": 1025220847,
35
+ "train_mem_gpu_alloc_delta": 3609088512,
36
+ "train_mem_gpu_peaked_delta": 8271511040,
37
+ "train_runtime": 64951.4124,
38
+ "train_samples": 266367,
39
+ "train_samples_per_second": 0.897
40
+ }
config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "LeoCordoba/mt5-small-mlsum",
3
+ "architectures": [
4
+ "MT5ForConditionalGeneration"
5
+ ],
6
+ "d_ff": 1024,
7
+ "d_kv": 64,
8
+ "d_model": 512,
9
+ "decoder_start_token_id": 0,
10
+ "dropout_rate": 0.1,
11
+ "eos_token_id": 1,
12
+ "feed_forward_proj": "gated-gelu",
13
+ "initializer_factor": 1.0,
14
+ "is_encoder_decoder": true,
15
+ "layer_norm_epsilon": 1e-06,
16
+ "model_type": "mt5",
17
+ "num_decoder_layers": 8,
18
+ "num_heads": 6,
19
+ "num_layers": 8,
20
+ "pad_token_id": 0,
21
+ "relative_attention_num_buckets": 32,
22
+ "tie_word_embeddings": false,
23
+ "tokenizer_class": "T5Tokenizer",
24
+ "transformers_version": "4.4.2",
25
+ "use_cache": true,
26
+ "vocab_size": 250112
27
+ }
eval_results.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 7.0,
3
+ "eval_gen_len": 33.9532,
4
+ "eval_loss": 1.9251619577407837,
5
+ "eval_mem_cpu_alloc_delta": 2314445767,
6
+ "eval_mem_cpu_peaked_delta": 26998734,
7
+ "eval_mem_gpu_alloc_delta": 0,
8
+ "eval_mem_gpu_peaked_delta": 3355142144,
9
+ "eval_rouge1": 26.4352,
10
+ "eval_rouge2": 8.9293,
11
+ "eval_rougeL": 21.2622,
12
+ "eval_rougeLsum": 21.5518,
13
+ "eval_runtime": 1054.0821,
14
+ "eval_samples": 10358,
15
+ "eval_samples_per_second": 9.827
16
+ }
metadata.json ADDED
@@ -0,0 +1 @@
 
 
1
+ [{"training-run": 0, "latest-global-step-saved": 8000, "latest-global-step-seen": 8323, "latest-mode-step": {"GLOBAL": 8323}}, {"training-run": 0, "latest-global-step-saved": 8000, "latest-global-step-seen": 8324, "latest-mode-step": {"GLOBAL": 8324}}, {"training-run": 0, "latest-global-step-saved": 8000, "latest-global-step-seen": 8325, "latest-mode-step": {"GLOBAL": 8325}}, {"training-run": 0, "latest-global-step-saved": 8000, "latest-global-step-seen": 8343, "latest-mode-step": {"GLOBAL": 8343}}, {"training-run": 0, "latest-global-step-saved": 16500, "latest-global-step-seen": 16647, "latest-mode-step": {"GLOBAL": 16647}}, {"training-run": 0, "latest-global-step-saved": 16500, "latest-global-step-seen": 16648, "latest-mode-step": {"GLOBAL": 16648}}, {"training-run": 0, "latest-global-step-saved": 16500, "latest-global-step-seen": 16669, "latest-mode-step": {"GLOBAL": 16669}}, {"training-run": 0, "latest-global-step-saved": 24500, "latest-global-step-seen": 24971, "latest-mode-step": {"GLOBAL": 24971}}, {"training-run": 0, "latest-global-step-saved": 24500, "latest-global-step-seen": 24972, "latest-mode-step": {"GLOBAL": 24972}}, {"training-run": 0, "latest-global-step-saved": 24500, "latest-global-step-seen": 24989, "latest-mode-step": {"GLOBAL": 24989}}, {"training-run": 0, "latest-global-step-saved": 33000, "latest-global-step-seen": 33295, "latest-mode-step": {"GLOBAL": 33295}}, {"training-run": 0, "latest-global-step-saved": 33000, "latest-global-step-seen": 33296, "latest-mode-step": {"GLOBAL": 33296}}, {"training-run": 0, "latest-global-step-saved": 33000, "latest-global-step-seen": 33318, "latest-mode-step": {"GLOBAL": 33318}}, {"training-run": 0, "latest-global-step-saved": 41500, "latest-global-step-seen": 41619, "latest-mode-step": {"GLOBAL": 41619}}, {"training-run": 0, "latest-global-step-saved": 41500, "latest-global-step-seen": 41620, "latest-mode-step": {"GLOBAL": 41620}}, {"training-run": 0, "latest-global-step-saved": 41500, "latest-global-step-seen": 41640, "latest-mode-step": {"GLOBAL": 41640}}, {"training-run": 0, "latest-global-step-saved": 49500, "latest-global-step-seen": 49943, "latest-mode-step": {"GLOBAL": 49943}}, {"training-run": 0, "latest-global-step-saved": 49500, "latest-global-step-seen": 49944, "latest-mode-step": {"GLOBAL": 49944}}, {"training-run": 0, "latest-global-step-saved": 49500, "latest-global-step-seen": 49964, "latest-mode-step": {"GLOBAL": 49964}}]
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4685a5d2dd964b1d4b17aa216f5c93a518328e5a4040b8a1a52eb3cd105addc9
3
+ size 1200775901
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>"}
spiece.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ef78f86560d809067d12bac6c09f19a462cb3af3f54d2b8acbba26e1433125d6
3
+ size 4309802
test_generations.txt ADDED
The diff for this file is too large to render. See raw diff
 
test_results.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "test_gen_len": 33.803,
3
+ "test_loss": 2.005567789077759,
4
+ "test_mem_cpu_alloc_delta": 3104676530,
5
+ "test_mem_cpu_peaked_delta": 36246285,
6
+ "test_mem_gpu_alloc_delta": 0,
7
+ "test_mem_gpu_peaked_delta": 3357975040,
8
+ "test_rouge1": 26.0756,
9
+ "test_rouge2": 8.4669,
10
+ "test_rougeL": 20.8167,
11
+ "test_rougeLsum": 21.0822,
12
+ "test_runtime": 1405.8563,
13
+ "test_samples": 13920,
14
+ "test_samples_per_second": 9.901
15
+ }
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>", "extra_ids": 0, "additional_special_tokens": null, "special_tokens_map_file": "/root/.cache/huggingface/transformers/685ac0ca8568ec593a48b61b0a3c272beee9bc194a3c7241d15dcadb5f875e53.f76030f3ec1b96a8199b2593390c610e76ca8028ef3d24680000619ffb646276", "name_or_path": "LeoCordoba/mt5-small-mlsum"}
train_results.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 7.0,
3
+ "init_mem_cpu_alloc_delta": 177892,
4
+ "init_mem_cpu_peaked_delta": 18282,
5
+ "init_mem_gpu_alloc_delta": 1200707584,
6
+ "init_mem_gpu_peaked_delta": 0,
7
+ "train_mem_cpu_alloc_delta": 9871729,
8
+ "train_mem_cpu_peaked_delta": 1025220847,
9
+ "train_mem_gpu_alloc_delta": 3609088512,
10
+ "train_mem_gpu_peaked_delta": 8271511040,
11
+ "train_runtime": 64951.4124,
12
+ "train_samples": 266367,
13
+ "train_samples_per_second": 0.897
14
+ }
trainer_state.json ADDED
@@ -0,0 +1,719 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 7.0,
5
+ "global_step": 58268,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.06,
12
+ "learning_rate": 4.957094803322579e-05,
13
+ "loss": 2.4741,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 0.12,
18
+ "learning_rate": 4.914189606645157e-05,
19
+ "loss": 2.4661,
20
+ "step": 1000
21
+ },
22
+ {
23
+ "epoch": 0.18,
24
+ "learning_rate": 4.871284409967736e-05,
25
+ "loss": 2.4505,
26
+ "step": 1500
27
+ },
28
+ {
29
+ "epoch": 0.24,
30
+ "learning_rate": 4.828379213290314e-05,
31
+ "loss": 2.4463,
32
+ "step": 2000
33
+ },
34
+ {
35
+ "epoch": 0.3,
36
+ "learning_rate": 4.785474016612893e-05,
37
+ "loss": 2.4309,
38
+ "step": 2500
39
+ },
40
+ {
41
+ "epoch": 0.36,
42
+ "learning_rate": 4.742568819935471e-05,
43
+ "loss": 2.4286,
44
+ "step": 3000
45
+ },
46
+ {
47
+ "epoch": 0.42,
48
+ "learning_rate": 4.699663623258049e-05,
49
+ "loss": 2.4326,
50
+ "step": 3500
51
+ },
52
+ {
53
+ "epoch": 0.48,
54
+ "learning_rate": 4.656758426580627e-05,
55
+ "loss": 2.424,
56
+ "step": 4000
57
+ },
58
+ {
59
+ "epoch": 0.54,
60
+ "learning_rate": 4.613853229903206e-05,
61
+ "loss": 2.4179,
62
+ "step": 4500
63
+ },
64
+ {
65
+ "epoch": 0.6,
66
+ "learning_rate": 4.570948033225785e-05,
67
+ "loss": 2.4141,
68
+ "step": 5000
69
+ },
70
+ {
71
+ "epoch": 0.66,
72
+ "learning_rate": 4.528042836548363e-05,
73
+ "loss": 2.4121,
74
+ "step": 5500
75
+ },
76
+ {
77
+ "epoch": 0.72,
78
+ "learning_rate": 4.4851376398709416e-05,
79
+ "loss": 2.407,
80
+ "step": 6000
81
+ },
82
+ {
83
+ "epoch": 0.78,
84
+ "learning_rate": 4.44223244319352e-05,
85
+ "loss": 2.399,
86
+ "step": 6500
87
+ },
88
+ {
89
+ "epoch": 0.84,
90
+ "learning_rate": 4.3993272465160985e-05,
91
+ "loss": 2.3935,
92
+ "step": 7000
93
+ },
94
+ {
95
+ "epoch": 0.9,
96
+ "learning_rate": 4.3564220498386766e-05,
97
+ "loss": 2.3822,
98
+ "step": 7500
99
+ },
100
+ {
101
+ "epoch": 0.96,
102
+ "learning_rate": 4.3135168531612554e-05,
103
+ "loss": 2.3822,
104
+ "step": 8000
105
+ },
106
+ {
107
+ "epoch": 1.02,
108
+ "learning_rate": 4.2706116564838335e-05,
109
+ "loss": 2.387,
110
+ "step": 8500
111
+ },
112
+ {
113
+ "epoch": 1.08,
114
+ "learning_rate": 4.227706459806412e-05,
115
+ "loss": 2.3601,
116
+ "step": 9000
117
+ },
118
+ {
119
+ "epoch": 1.14,
120
+ "learning_rate": 4.1848012631289904e-05,
121
+ "loss": 2.3582,
122
+ "step": 9500
123
+ },
124
+ {
125
+ "epoch": 1.2,
126
+ "learning_rate": 4.1418960664515685e-05,
127
+ "loss": 2.3706,
128
+ "step": 10000
129
+ },
130
+ {
131
+ "epoch": 1.26,
132
+ "learning_rate": 4.098990869774147e-05,
133
+ "loss": 2.3669,
134
+ "step": 10500
135
+ },
136
+ {
137
+ "epoch": 1.32,
138
+ "learning_rate": 4.0560856730967254e-05,
139
+ "loss": 2.3518,
140
+ "step": 11000
141
+ },
142
+ {
143
+ "epoch": 1.38,
144
+ "learning_rate": 4.013180476419304e-05,
145
+ "loss": 2.3392,
146
+ "step": 11500
147
+ },
148
+ {
149
+ "epoch": 1.44,
150
+ "learning_rate": 3.970275279741882e-05,
151
+ "loss": 2.3502,
152
+ "step": 12000
153
+ },
154
+ {
155
+ "epoch": 1.5,
156
+ "learning_rate": 3.927370083064461e-05,
157
+ "loss": 2.3437,
158
+ "step": 12500
159
+ },
160
+ {
161
+ "epoch": 1.56,
162
+ "learning_rate": 3.884464886387039e-05,
163
+ "loss": 2.3577,
164
+ "step": 13000
165
+ },
166
+ {
167
+ "epoch": 1.62,
168
+ "learning_rate": 3.841559689709618e-05,
169
+ "loss": 2.3435,
170
+ "step": 13500
171
+ },
172
+ {
173
+ "epoch": 1.68,
174
+ "learning_rate": 3.798654493032196e-05,
175
+ "loss": 2.3456,
176
+ "step": 14000
177
+ },
178
+ {
179
+ "epoch": 1.74,
180
+ "learning_rate": 3.755749296354775e-05,
181
+ "loss": 2.3461,
182
+ "step": 14500
183
+ },
184
+ {
185
+ "epoch": 1.8,
186
+ "learning_rate": 3.712844099677354e-05,
187
+ "loss": 2.3472,
188
+ "step": 15000
189
+ },
190
+ {
191
+ "epoch": 1.86,
192
+ "learning_rate": 3.669938902999932e-05,
193
+ "loss": 2.3362,
194
+ "step": 15500
195
+ },
196
+ {
197
+ "epoch": 1.92,
198
+ "learning_rate": 3.62703370632251e-05,
199
+ "loss": 2.345,
200
+ "step": 16000
201
+ },
202
+ {
203
+ "epoch": 1.98,
204
+ "learning_rate": 3.584128509645088e-05,
205
+ "loss": 2.336,
206
+ "step": 16500
207
+ },
208
+ {
209
+ "epoch": 2.04,
210
+ "learning_rate": 3.541223312967667e-05,
211
+ "loss": 2.3365,
212
+ "step": 17000
213
+ },
214
+ {
215
+ "epoch": 2.1,
216
+ "learning_rate": 3.498318116290245e-05,
217
+ "loss": 2.3245,
218
+ "step": 17500
219
+ },
220
+ {
221
+ "epoch": 2.16,
222
+ "learning_rate": 3.455412919612824e-05,
223
+ "loss": 2.3086,
224
+ "step": 18000
225
+ },
226
+ {
227
+ "epoch": 2.22,
228
+ "learning_rate": 3.412507722935402e-05,
229
+ "loss": 2.3251,
230
+ "step": 18500
231
+ },
232
+ {
233
+ "epoch": 2.28,
234
+ "learning_rate": 3.3696025262579806e-05,
235
+ "loss": 2.32,
236
+ "step": 19000
237
+ },
238
+ {
239
+ "epoch": 2.34,
240
+ "learning_rate": 3.326697329580559e-05,
241
+ "loss": 2.3084,
242
+ "step": 19500
243
+ },
244
+ {
245
+ "epoch": 2.4,
246
+ "learning_rate": 3.2837921329031375e-05,
247
+ "loss": 2.3164,
248
+ "step": 20000
249
+ },
250
+ {
251
+ "epoch": 2.46,
252
+ "learning_rate": 3.240886936225716e-05,
253
+ "loss": 2.3054,
254
+ "step": 20500
255
+ },
256
+ {
257
+ "epoch": 2.52,
258
+ "learning_rate": 3.1979817395482944e-05,
259
+ "loss": 2.3172,
260
+ "step": 21000
261
+ },
262
+ {
263
+ "epoch": 2.58,
264
+ "learning_rate": 3.1550765428708725e-05,
265
+ "loss": 2.31,
266
+ "step": 21500
267
+ },
268
+ {
269
+ "epoch": 2.64,
270
+ "learning_rate": 3.112171346193451e-05,
271
+ "loss": 2.3141,
272
+ "step": 22000
273
+ },
274
+ {
275
+ "epoch": 2.7,
276
+ "learning_rate": 3.0692661495160294e-05,
277
+ "loss": 2.3226,
278
+ "step": 22500
279
+ },
280
+ {
281
+ "epoch": 2.76,
282
+ "learning_rate": 3.026360952838608e-05,
283
+ "loss": 2.3194,
284
+ "step": 23000
285
+ },
286
+ {
287
+ "epoch": 2.82,
288
+ "learning_rate": 2.9834557561611863e-05,
289
+ "loss": 2.3203,
290
+ "step": 23500
291
+ },
292
+ {
293
+ "epoch": 2.88,
294
+ "learning_rate": 2.9405505594837644e-05,
295
+ "loss": 2.3065,
296
+ "step": 24000
297
+ },
298
+ {
299
+ "epoch": 2.94,
300
+ "learning_rate": 2.8976453628063432e-05,
301
+ "loss": 2.3112,
302
+ "step": 24500
303
+ },
304
+ {
305
+ "epoch": 3.0,
306
+ "learning_rate": 2.854740166128922e-05,
307
+ "loss": 2.2985,
308
+ "step": 25000
309
+ },
310
+ {
311
+ "epoch": 3.06,
312
+ "learning_rate": 2.8118349694515e-05,
313
+ "loss": 2.3152,
314
+ "step": 25500
315
+ },
316
+ {
317
+ "epoch": 3.12,
318
+ "learning_rate": 2.768929772774079e-05,
319
+ "loss": 2.2912,
320
+ "step": 26000
321
+ },
322
+ {
323
+ "epoch": 3.18,
324
+ "learning_rate": 2.726024576096657e-05,
325
+ "loss": 2.2759,
326
+ "step": 26500
327
+ },
328
+ {
329
+ "epoch": 3.24,
330
+ "learning_rate": 2.6831193794192354e-05,
331
+ "loss": 2.2952,
332
+ "step": 27000
333
+ },
334
+ {
335
+ "epoch": 3.3,
336
+ "learning_rate": 2.6402141827418136e-05,
337
+ "loss": 2.2971,
338
+ "step": 27500
339
+ },
340
+ {
341
+ "epoch": 3.36,
342
+ "learning_rate": 2.5973089860643923e-05,
343
+ "loss": 2.2987,
344
+ "step": 28000
345
+ },
346
+ {
347
+ "epoch": 3.42,
348
+ "learning_rate": 2.5544037893869704e-05,
349
+ "loss": 2.2859,
350
+ "step": 28500
351
+ },
352
+ {
353
+ "epoch": 3.48,
354
+ "learning_rate": 2.5114985927095492e-05,
355
+ "loss": 2.2779,
356
+ "step": 29000
357
+ },
358
+ {
359
+ "epoch": 3.54,
360
+ "learning_rate": 2.4685933960321277e-05,
361
+ "loss": 2.2878,
362
+ "step": 29500
363
+ },
364
+ {
365
+ "epoch": 3.6,
366
+ "learning_rate": 2.4256881993547058e-05,
367
+ "loss": 2.2762,
368
+ "step": 30000
369
+ },
370
+ {
371
+ "epoch": 3.66,
372
+ "learning_rate": 2.3827830026772842e-05,
373
+ "loss": 2.2728,
374
+ "step": 30500
375
+ },
376
+ {
377
+ "epoch": 3.72,
378
+ "learning_rate": 2.3398778059998627e-05,
379
+ "loss": 2.2898,
380
+ "step": 31000
381
+ },
382
+ {
383
+ "epoch": 3.78,
384
+ "learning_rate": 2.296972609322441e-05,
385
+ "loss": 2.2836,
386
+ "step": 31500
387
+ },
388
+ {
389
+ "epoch": 3.84,
390
+ "learning_rate": 2.25406741264502e-05,
391
+ "loss": 2.2882,
392
+ "step": 32000
393
+ },
394
+ {
395
+ "epoch": 3.9,
396
+ "learning_rate": 2.211162215967598e-05,
397
+ "loss": 2.2778,
398
+ "step": 32500
399
+ },
400
+ {
401
+ "epoch": 3.96,
402
+ "learning_rate": 2.1682570192901765e-05,
403
+ "loss": 2.2793,
404
+ "step": 33000
405
+ },
406
+ {
407
+ "epoch": 4.02,
408
+ "learning_rate": 2.125351822612755e-05,
409
+ "loss": 2.2699,
410
+ "step": 33500
411
+ },
412
+ {
413
+ "epoch": 4.08,
414
+ "learning_rate": 2.0824466259353334e-05,
415
+ "loss": 2.2778,
416
+ "step": 34000
417
+ },
418
+ {
419
+ "epoch": 4.14,
420
+ "learning_rate": 2.0395414292579118e-05,
421
+ "loss": 2.2668,
422
+ "step": 34500
423
+ },
424
+ {
425
+ "epoch": 4.2,
426
+ "learning_rate": 1.9966362325804903e-05,
427
+ "loss": 2.257,
428
+ "step": 35000
429
+ },
430
+ {
431
+ "epoch": 4.26,
432
+ "learning_rate": 1.9537310359030687e-05,
433
+ "loss": 2.2496,
434
+ "step": 35500
435
+ },
436
+ {
437
+ "epoch": 4.32,
438
+ "learning_rate": 1.9108258392256472e-05,
439
+ "loss": 2.2742,
440
+ "step": 36000
441
+ },
442
+ {
443
+ "epoch": 4.38,
444
+ "learning_rate": 1.8679206425482253e-05,
445
+ "loss": 2.2643,
446
+ "step": 36500
447
+ },
448
+ {
449
+ "epoch": 4.44,
450
+ "learning_rate": 1.825015445870804e-05,
451
+ "loss": 2.2517,
452
+ "step": 37000
453
+ },
454
+ {
455
+ "epoch": 4.51,
456
+ "learning_rate": 1.7821102491933825e-05,
457
+ "loss": 2.2677,
458
+ "step": 37500
459
+ },
460
+ {
461
+ "epoch": 4.57,
462
+ "learning_rate": 1.739205052515961e-05,
463
+ "loss": 2.2654,
464
+ "step": 38000
465
+ },
466
+ {
467
+ "epoch": 4.63,
468
+ "learning_rate": 1.6962998558385394e-05,
469
+ "loss": 2.2645,
470
+ "step": 38500
471
+ },
472
+ {
473
+ "epoch": 4.69,
474
+ "learning_rate": 1.6533946591611175e-05,
475
+ "loss": 2.2456,
476
+ "step": 39000
477
+ },
478
+ {
479
+ "epoch": 4.75,
480
+ "learning_rate": 1.610489462483696e-05,
481
+ "loss": 2.2686,
482
+ "step": 39500
483
+ },
484
+ {
485
+ "epoch": 4.81,
486
+ "learning_rate": 1.5675842658062744e-05,
487
+ "loss": 2.2478,
488
+ "step": 40000
489
+ },
490
+ {
491
+ "epoch": 4.87,
492
+ "learning_rate": 1.5246790691288529e-05,
493
+ "loss": 2.2558,
494
+ "step": 40500
495
+ },
496
+ {
497
+ "epoch": 4.93,
498
+ "learning_rate": 1.4817738724514313e-05,
499
+ "loss": 2.2541,
500
+ "step": 41000
501
+ },
502
+ {
503
+ "epoch": 4.99,
504
+ "learning_rate": 1.4388686757740098e-05,
505
+ "loss": 2.236,
506
+ "step": 41500
507
+ },
508
+ {
509
+ "epoch": 5.05,
510
+ "learning_rate": 1.3959634790965884e-05,
511
+ "loss": 2.2543,
512
+ "step": 42000
513
+ },
514
+ {
515
+ "epoch": 5.11,
516
+ "learning_rate": 1.3530582824191668e-05,
517
+ "loss": 2.2479,
518
+ "step": 42500
519
+ },
520
+ {
521
+ "epoch": 5.17,
522
+ "learning_rate": 1.3101530857417451e-05,
523
+ "loss": 2.2525,
524
+ "step": 43000
525
+ },
526
+ {
527
+ "epoch": 5.23,
528
+ "learning_rate": 1.2672478890643236e-05,
529
+ "loss": 2.2506,
530
+ "step": 43500
531
+ },
532
+ {
533
+ "epoch": 5.29,
534
+ "learning_rate": 1.224342692386902e-05,
535
+ "loss": 2.2282,
536
+ "step": 44000
537
+ },
538
+ {
539
+ "epoch": 5.35,
540
+ "learning_rate": 1.1814374957094803e-05,
541
+ "loss": 2.2406,
542
+ "step": 44500
543
+ },
544
+ {
545
+ "epoch": 5.41,
546
+ "learning_rate": 1.1385322990320587e-05,
547
+ "loss": 2.2357,
548
+ "step": 45000
549
+ },
550
+ {
551
+ "epoch": 5.47,
552
+ "learning_rate": 1.0956271023546374e-05,
553
+ "loss": 2.2457,
554
+ "step": 45500
555
+ },
556
+ {
557
+ "epoch": 5.53,
558
+ "learning_rate": 1.0527219056772156e-05,
559
+ "loss": 2.2262,
560
+ "step": 46000
561
+ },
562
+ {
563
+ "epoch": 5.59,
564
+ "learning_rate": 1.0098167089997941e-05,
565
+ "loss": 2.2423,
566
+ "step": 46500
567
+ },
568
+ {
569
+ "epoch": 5.65,
570
+ "learning_rate": 9.669115123223725e-06,
571
+ "loss": 2.2435,
572
+ "step": 47000
573
+ },
574
+ {
575
+ "epoch": 5.71,
576
+ "learning_rate": 9.24006315644951e-06,
577
+ "loss": 2.2277,
578
+ "step": 47500
579
+ },
580
+ {
581
+ "epoch": 5.77,
582
+ "learning_rate": 8.811011189675294e-06,
583
+ "loss": 2.2445,
584
+ "step": 48000
585
+ },
586
+ {
587
+ "epoch": 5.83,
588
+ "learning_rate": 8.381959222901079e-06,
589
+ "loss": 2.2395,
590
+ "step": 48500
591
+ },
592
+ {
593
+ "epoch": 5.89,
594
+ "learning_rate": 7.952907256126863e-06,
595
+ "loss": 2.2506,
596
+ "step": 49000
597
+ },
598
+ {
599
+ "epoch": 5.95,
600
+ "learning_rate": 7.523855289352647e-06,
601
+ "loss": 2.234,
602
+ "step": 49500
603
+ },
604
+ {
605
+ "epoch": 6.01,
606
+ "learning_rate": 7.0948033225784306e-06,
607
+ "loss": 2.2488,
608
+ "step": 50000
609
+ },
610
+ {
611
+ "epoch": 6.07,
612
+ "learning_rate": 6.665751355804216e-06,
613
+ "loss": 2.2242,
614
+ "step": 50500
615
+ },
616
+ {
617
+ "epoch": 6.13,
618
+ "learning_rate": 6.2366993890299995e-06,
619
+ "loss": 2.2297,
620
+ "step": 51000
621
+ },
622
+ {
623
+ "epoch": 6.19,
624
+ "learning_rate": 5.807647422255784e-06,
625
+ "loss": 2.2259,
626
+ "step": 51500
627
+ },
628
+ {
629
+ "epoch": 6.25,
630
+ "learning_rate": 5.378595455481568e-06,
631
+ "loss": 2.2263,
632
+ "step": 52000
633
+ },
634
+ {
635
+ "epoch": 6.31,
636
+ "learning_rate": 4.949543488707353e-06,
637
+ "loss": 2.2214,
638
+ "step": 52500
639
+ },
640
+ {
641
+ "epoch": 6.37,
642
+ "learning_rate": 4.520491521933137e-06,
643
+ "loss": 2.2287,
644
+ "step": 53000
645
+ },
646
+ {
647
+ "epoch": 6.43,
648
+ "learning_rate": 4.091439555158921e-06,
649
+ "loss": 2.2384,
650
+ "step": 53500
651
+ },
652
+ {
653
+ "epoch": 6.49,
654
+ "learning_rate": 3.662387588384705e-06,
655
+ "loss": 2.2319,
656
+ "step": 54000
657
+ },
658
+ {
659
+ "epoch": 6.55,
660
+ "learning_rate": 3.2333356216104892e-06,
661
+ "loss": 2.2328,
662
+ "step": 54500
663
+ },
664
+ {
665
+ "epoch": 6.61,
666
+ "learning_rate": 2.804283654836274e-06,
667
+ "loss": 2.2359,
668
+ "step": 55000
669
+ },
670
+ {
671
+ "epoch": 6.67,
672
+ "learning_rate": 2.375231688062058e-06,
673
+ "loss": 2.2372,
674
+ "step": 55500
675
+ },
676
+ {
677
+ "epoch": 6.73,
678
+ "learning_rate": 1.9461797212878423e-06,
679
+ "loss": 2.2402,
680
+ "step": 56000
681
+ },
682
+ {
683
+ "epoch": 6.79,
684
+ "learning_rate": 1.5171277545136267e-06,
685
+ "loss": 2.2243,
686
+ "step": 56500
687
+ },
688
+ {
689
+ "epoch": 6.85,
690
+ "learning_rate": 1.088075787739411e-06,
691
+ "loss": 2.2267,
692
+ "step": 57000
693
+ },
694
+ {
695
+ "epoch": 6.91,
696
+ "learning_rate": 6.590238209651953e-07,
697
+ "loss": 2.2254,
698
+ "step": 57500
699
+ },
700
+ {
701
+ "epoch": 6.97,
702
+ "learning_rate": 2.2997185419097963e-07,
703
+ "loss": 2.2278,
704
+ "step": 58000
705
+ },
706
+ {
707
+ "epoch": 7.0,
708
+ "step": 58268,
709
+ "total_flos": 3.6314933518540524e+18,
710
+ "train_runtime": 64951.4124,
711
+ "train_samples_per_second": 0.897
712
+ }
713
+ ],
714
+ "max_steps": 58268,
715
+ "num_train_epochs": 7,
716
+ "total_flos": 3.6314933518540524e+18,
717
+ "trial_name": null,
718
+ "trial_params": null
719
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5a06ccbebdbae67cab496e1bec18b38558efc6e9c7bf93f835222fe5f03dccf9
3
+ size 2351