Hugging Face's logo

Langame
/
gpt2-starter

Text Generation
Transformers
PyTorch
TensorBoard
ONNX
gpt2
text-generation-inference
Inference Endpoints
Model card Files Metrics Community
louis030195 commited on
Commit
4416589
•
1 Parent(s): d876e40

update

Browse files
Files changed (19) hide show
  1. .gitignore +2 -0
  2. all_results.json +14 -0
  3. config.json +4 -6
  4. eval_results.json +9 -0
  5. merges.txt +0 -0
  6. onnx/model.onnx +3 -0
  7. pytorch_model.bin +2 -2
  8. runs/Dec11_12-09-41_louis-gpu/events.out.tfevents.1639221003.louis-gpu +3 -0
  9. runs/Dec11_12-09-41_louis-gpu/events.out.tfevents.1639221263.louis-gpu +3 -0
  10. runs/Dec11_12-26-07_louis-gpu/1639221986.8950253/events.out.tfevents.1639221986.louis-gpu.1530706.1 +3 -0
  11. runs/Dec11_12-26-07_louis-gpu/events.out.tfevents.1639221986.louis-gpu.1530706.0 +3 -0
  12. runs/Dec11_17-59-34_louis-gpu/1639241995.1505647/events.out.tfevents.1639241995.louis-gpu.3102814.1 +3 -0
  13. runs/Dec11_17-59-34_louis-gpu/events.out.tfevents.1639241995.louis-gpu.3102814.0 +3 -0
  14. runs/Dec11_17-59-34_louis-gpu/events.out.tfevents.1639250844.louis-gpu.3102814.2 +3 -0
  15. special_tokens_map.json +1 -0
  16. tokenizer_config.json +1 -0
  17. train_results.json +8 -0
  18. trainer_state.json +925 -0
  19. training_args.bin +3 -0
.gitignore ADDED
@@ -0,0 +1,2 @@
 
 
 
1
+ checkpoint*
2
+ wandb
all_results.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 100.0,
3
+ "eval_loss": 5.671875,
4
+ "eval_runtime": 11.7689,
5
+ "eval_samples": 801,
6
+ "eval_samples_per_second": 68.061,
7
+ "eval_steps_per_second": 8.582,
8
+ "perplexity": 290.5788592888401,
9
+ "train_loss": 0.3251776333512931,
10
+ "train_runtime": 8837.2329,
11
+ "train_samples": 2792,
12
+ "train_samples_per_second": 31.594,
13
+ "train_steps_per_second": 1.969
14
+ }
config.json CHANGED
@@ -1,13 +1,14 @@
1
  {
2
- "_name_or_path": "./finetuned",
3
  "activation_function": "gelu_new",
4
  "architectures": [
5
- "GPT2Model"
6
  ],
7
  "attn_pdrop": 0.1,
8
  "bos_token_id": 50256,
9
  "embd_pdrop": 0.1,
10
  "eos_token_id": 50256,
 
11
  "initializer_range": 0.02,
12
  "layer_norm_epsilon": 1e-05,
13
  "model_type": "gpt2",
@@ -17,9 +18,7 @@
17
  "n_inner": null,
18
  "n_layer": 12,
19
  "n_positions": 1024,
20
- "reorder_and_upcast_attn": false,
21
  "resid_pdrop": 0.1,
22
- "scale_attn_by_inverse_layer_idx": false,
23
  "scale_attn_weights": true,
24
  "summary_activation": null,
25
  "summary_first_dropout": 0.1,
@@ -32,8 +31,7 @@
32
  "max_length": 50
33
  }
34
  },
35
- "torch_dtype": "float32",
36
- "transformers_version": "4.13.0.dev0",
37
  "use_cache": false,
38
  "vocab_size": 50257
39
  }
 
1
  {
2
+ "_name_or_path": "gpt2",
3
  "activation_function": "gelu_new",
4
  "architectures": [
5
+ "GPT2LMHeadModel"
6
  ],
7
  "attn_pdrop": 0.1,
8
  "bos_token_id": 50256,
9
  "embd_pdrop": 0.1,
10
  "eos_token_id": 50256,
11
+ "gradient_checkpointing": true,
12
  "initializer_range": 0.02,
13
  "layer_norm_epsilon": 1e-05,
14
  "model_type": "gpt2",
 
18
  "n_inner": null,
19
  "n_layer": 12,
20
  "n_positions": 1024,
 
21
  "resid_pdrop": 0.1,
 
22
  "scale_attn_weights": true,
23
  "summary_activation": null,
24
  "summary_first_dropout": 0.1,
 
31
  "max_length": 50
32
  }
33
  },
34
+ "transformers_version": "4.7.0",
 
35
  "use_cache": false,
36
  "vocab_size": 50257
37
  }
eval_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 100.0,
3
+ "eval_loss": 5.671875,
4
+ "eval_runtime": 11.7689,
5
+ "eval_samples": 801,
6
+ "eval_samples_per_second": 68.061,
7
+ "eval_steps_per_second": 8.582,
8
+ "perplexity": 290.5788592888401
9
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
onnx/model.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6e19733f0bd626cd5aeeff42d82419c4407622a18f882aa860c832e20a699127
3
+ size 510470587
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0c2bab2b1ce984d2e24b2c59acc33354b61cd294a3287e5859d5077be5bc6906
3
- size 510397417
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7e25804ffb594892b6b6db2367e5e19c25ba50fe952d6abc0368e77b55f3168e
3
+ size 261495703
runs/Dec11_12-09-41_louis-gpu/events.out.tfevents.1639221003.louis-gpu ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8e5847de0185c670aa627e037c414069ced7793d24dbbbeb55a63e2ad16356c9
3
+ size 3313
runs/Dec11_12-09-41_louis-gpu/events.out.tfevents.1639221263.louis-gpu ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c831d79cafee548b62059cb8e43da123643cded0af6d3138c6a3d4073f9ba8a5
3
+ size 311
runs/Dec11_12-26-07_louis-gpu/1639221986.8950253/events.out.tfevents.1639221986.louis-gpu.1530706.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:88d03fba1a3d306c623da909a8f568082f9b878e55a2b2055d53a23d4de789fa
3
+ size 3852
runs/Dec11_12-26-07_louis-gpu/events.out.tfevents.1639221986.louis-gpu.1530706.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bc5c8170f82eface63fb642035f3d6a0dfd5d5f6d0fac73090fd7f15066b1890
3
+ size 22561
runs/Dec11_17-59-34_louis-gpu/1639241995.1505647/events.out.tfevents.1639241995.louis-gpu.3102814.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4dfbe2a4fe94b962578d509239b447f10759ad29c5352c1cde0fad925a30db18
3
+ size 3852
runs/Dec11_17-59-34_louis-gpu/events.out.tfevents.1639241995.louis-gpu.3102814.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0e15f05a8b70f8f372bfcdad617db5039dce4e6f86fa54c4d960f0f81a37fabf
3
+ size 13214
runs/Dec11_17-59-34_louis-gpu/events.out.tfevents.1639250844.louis-gpu.3102814.2 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1cf23acbcb15b242a8ffa22233b9244ed8c2d7478fb1d5a23a1c8c468564c29b
3
+ size 316
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "unk_token": "<|endoftext|>"}
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"unk_token": "<|endoftext|>", "bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "add_prefix_space": false, "model_max_length": 1024, "special_tokens_map_file": null, "name_or_path": "gpt2"}
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 100.0,
3
+ "train_loss": 0.3251776333512931,
4
+ "train_runtime": 8837.2329,
5
+ "train_samples": 2792,
6
+ "train_samples_per_second": 31.594,
7
+ "train_steps_per_second": 1.969
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,925 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 99.9971346704871,
5
+ "global_step": 17400,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 1.15,
12
+ "eval_loss": 3.23046875,
13
+ "eval_runtime": 11.7347,
14
+ "eval_samples_per_second": 68.259,
15
+ "eval_steps_per_second": 8.607,
16
+ "step": 200
17
+ },
18
+ {
19
+ "epoch": 2.3,
20
+ "eval_loss": 3.255859375,
21
+ "eval_runtime": 11.7326,
22
+ "eval_samples_per_second": 68.271,
23
+ "eval_steps_per_second": 8.608,
24
+ "step": 400
25
+ },
26
+ {
27
+ "epoch": 2.87,
28
+ "learning_rate": 5e-05,
29
+ "loss": 2.0922,
30
+ "step": 500
31
+ },
32
+ {
33
+ "epoch": 3.45,
34
+ "eval_loss": 3.279296875,
35
+ "eval_runtime": 11.7456,
36
+ "eval_samples_per_second": 68.196,
37
+ "eval_steps_per_second": 8.599,
38
+ "step": 600
39
+ },
40
+ {
41
+ "epoch": 4.6,
42
+ "eval_loss": 3.294921875,
43
+ "eval_runtime": 11.7533,
44
+ "eval_samples_per_second": 68.151,
45
+ "eval_steps_per_second": 8.593,
46
+ "step": 800
47
+ },
48
+ {
49
+ "epoch": 5.74,
50
+ "learning_rate": 5e-05,
51
+ "loss": 1.931,
52
+ "step": 1000
53
+ },
54
+ {
55
+ "epoch": 5.74,
56
+ "eval_loss": 3.310546875,
57
+ "eval_runtime": 11.7529,
58
+ "eval_samples_per_second": 68.153,
59
+ "eval_steps_per_second": 8.594,
60
+ "step": 1000
61
+ },
62
+ {
63
+ "epoch": 6.89,
64
+ "eval_loss": 3.30078125,
65
+ "eval_runtime": 11.76,
66
+ "eval_samples_per_second": 68.112,
67
+ "eval_steps_per_second": 8.588,
68
+ "step": 1200
69
+ },
70
+ {
71
+ "epoch": 8.05,
72
+ "eval_loss": 3.3359375,
73
+ "eval_runtime": 11.7654,
74
+ "eval_samples_per_second": 68.081,
75
+ "eval_steps_per_second": 8.584,
76
+ "step": 1400
77
+ },
78
+ {
79
+ "epoch": 8.62,
80
+ "learning_rate": 5e-05,
81
+ "loss": 1.8589,
82
+ "step": 1500
83
+ },
84
+ {
85
+ "epoch": 9.19,
86
+ "eval_loss": 3.341796875,
87
+ "eval_runtime": 11.7717,
88
+ "eval_samples_per_second": 68.045,
89
+ "eval_steps_per_second": 8.58,
90
+ "step": 1600
91
+ },
92
+ {
93
+ "epoch": 10.34,
94
+ "eval_loss": 3.353515625,
95
+ "eval_runtime": 11.7648,
96
+ "eval_samples_per_second": 68.085,
97
+ "eval_steps_per_second": 8.585,
98
+ "step": 1800
99
+ },
100
+ {
101
+ "epoch": 11.49,
102
+ "learning_rate": 5e-05,
103
+ "loss": 1.7982,
104
+ "step": 2000
105
+ },
106
+ {
107
+ "epoch": 11.49,
108
+ "eval_loss": 3.373046875,
109
+ "eval_runtime": 11.771,
110
+ "eval_samples_per_second": 68.049,
111
+ "eval_steps_per_second": 8.58,
112
+ "step": 2000
113
+ },
114
+ {
115
+ "epoch": 12.64,
116
+ "eval_loss": 3.37109375,
117
+ "eval_runtime": 11.7674,
118
+ "eval_samples_per_second": 68.069,
119
+ "eval_steps_per_second": 8.583,
120
+ "step": 2200
121
+ },
122
+ {
123
+ "epoch": 13.79,
124
+ "eval_loss": 3.40625,
125
+ "eval_runtime": 11.7675,
126
+ "eval_samples_per_second": 68.069,
127
+ "eval_steps_per_second": 8.583,
128
+ "step": 2400
129
+ },
130
+ {
131
+ "epoch": 14.37,
132
+ "learning_rate": 5e-05,
133
+ "loss": 1.7446,
134
+ "step": 2500
135
+ },
136
+ {
137
+ "epoch": 14.94,
138
+ "eval_loss": 3.4140625,
139
+ "eval_runtime": 11.7686,
140
+ "eval_samples_per_second": 68.062,
141
+ "eval_steps_per_second": 8.582,
142
+ "step": 2600
143
+ },
144
+ {
145
+ "epoch": 16.09,
146
+ "eval_loss": 3.45703125,
147
+ "eval_runtime": 11.7612,
148
+ "eval_samples_per_second": 68.105,
149
+ "eval_steps_per_second": 8.588,
150
+ "step": 2800
151
+ },
152
+ {
153
+ "epoch": 17.24,
154
+ "learning_rate": 5e-05,
155
+ "loss": 1.6947,
156
+ "step": 3000
157
+ },
158
+ {
159
+ "epoch": 17.24,
160
+ "eval_loss": 3.498046875,
161
+ "eval_runtime": 11.7651,
162
+ "eval_samples_per_second": 68.082,
163
+ "eval_steps_per_second": 8.585,
164
+ "step": 3000
165
+ },
166
+ {
167
+ "epoch": 18.39,
168
+ "eval_loss": 3.51171875,
169
+ "eval_runtime": 11.7624,
170
+ "eval_samples_per_second": 68.098,
171
+ "eval_steps_per_second": 8.587,
172
+ "step": 3200
173
+ },
174
+ {
175
+ "epoch": 19.54,
176
+ "eval_loss": 3.533203125,
177
+ "eval_runtime": 11.7616,
178
+ "eval_samples_per_second": 68.103,
179
+ "eval_steps_per_second": 8.587,
180
+ "step": 3400
181
+ },
182
+ {
183
+ "epoch": 20.11,
184
+ "learning_rate": 5e-05,
185
+ "loss": 1.6464,
186
+ "step": 3500
187
+ },
188
+ {
189
+ "epoch": 20.69,
190
+ "eval_loss": 3.55078125,
191
+ "eval_runtime": 11.7586,
192
+ "eval_samples_per_second": 68.121,
193
+ "eval_steps_per_second": 8.589,
194
+ "step": 3600
195
+ },
196
+ {
197
+ "epoch": 21.84,
198
+ "eval_loss": 3.580078125,
199
+ "eval_runtime": 11.772,
200
+ "eval_samples_per_second": 68.043,
201
+ "eval_steps_per_second": 8.58,
202
+ "step": 3800
203
+ },
204
+ {
205
+ "epoch": 22.99,
206
+ "learning_rate": 5e-05,
207
+ "loss": 1.5981,
208
+ "step": 4000
209
+ },
210
+ {
211
+ "epoch": 22.99,
212
+ "eval_loss": 3.6015625,
213
+ "eval_runtime": 11.7661,
214
+ "eval_samples_per_second": 68.077,
215
+ "eval_steps_per_second": 8.584,
216
+ "step": 4000
217
+ },
218
+ {
219
+ "epoch": 24.14,
220
+ "eval_loss": 3.6796875,
221
+ "eval_runtime": 11.7648,
222
+ "eval_samples_per_second": 68.085,
223
+ "eval_steps_per_second": 8.585,
224
+ "step": 4200
225
+ },
226
+ {
227
+ "epoch": 25.29,
228
+ "eval_loss": 3.7265625,
229
+ "eval_runtime": 11.7655,
230
+ "eval_samples_per_second": 68.08,
231
+ "eval_steps_per_second": 8.584,
232
+ "step": 4400
233
+ },
234
+ {
235
+ "epoch": 25.86,
236
+ "learning_rate": 5e-05,
237
+ "loss": 1.5539,
238
+ "step": 4500
239
+ },
240
+ {
241
+ "epoch": 26.44,
242
+ "eval_loss": 3.736328125,
243
+ "eval_runtime": 11.7698,
244
+ "eval_samples_per_second": 68.056,
245
+ "eval_steps_per_second": 8.581,
246
+ "step": 4600
247
+ },
248
+ {
249
+ "epoch": 27.58,
250
+ "eval_loss": 3.755859375,
251
+ "eval_runtime": 11.7676,
252
+ "eval_samples_per_second": 68.068,
253
+ "eval_steps_per_second": 8.583,
254
+ "step": 4800
255
+ },
256
+ {
257
+ "epoch": 28.73,
258
+ "learning_rate": 5e-05,
259
+ "loss": 1.5105,
260
+ "step": 5000
261
+ },
262
+ {
263
+ "epoch": 28.73,
264
+ "eval_loss": 3.79296875,
265
+ "eval_runtime": 11.7657,
266
+ "eval_samples_per_second": 68.079,
267
+ "eval_steps_per_second": 8.584,
268
+ "step": 5000
269
+ },
270
+ {
271
+ "epoch": 29.88,
272
+ "eval_loss": 3.859375,
273
+ "eval_runtime": 11.7702,
274
+ "eval_samples_per_second": 68.053,
275
+ "eval_steps_per_second": 8.581,
276
+ "step": 5200
277
+ },
278
+ {
279
+ "epoch": 31.03,
280
+ "eval_loss": 3.951171875,
281
+ "eval_runtime": 11.7682,
282
+ "eval_samples_per_second": 68.065,
283
+ "eval_steps_per_second": 8.582,
284
+ "step": 5400
285
+ },
286
+ {
287
+ "epoch": 31.61,
288
+ "learning_rate": 5e-05,
289
+ "loss": 1.4699,
290
+ "step": 5500
291
+ },
292
+ {
293
+ "epoch": 32.18,
294
+ "eval_loss": 3.953125,
295
+ "eval_runtime": 11.768,
296
+ "eval_samples_per_second": 68.066,
297
+ "eval_steps_per_second": 8.583,
298
+ "step": 5600
299
+ },
300
+ {
301
+ "epoch": 33.33,
302
+ "eval_loss": 3.95703125,
303
+ "eval_runtime": 11.7717,
304
+ "eval_samples_per_second": 68.044,
305
+ "eval_steps_per_second": 8.58,
306
+ "step": 5800
307
+ },
308
+ {
309
+ "epoch": 34.48,
310
+ "learning_rate": 5e-05,
311
+ "loss": 1.4317,
312
+ "step": 6000
313
+ },
314
+ {
315
+ "epoch": 34.48,
316
+ "eval_loss": 4.0234375,
317
+ "eval_runtime": 11.7662,
318
+ "eval_samples_per_second": 68.076,
319
+ "eval_steps_per_second": 8.584,
320
+ "step": 6000
321
+ },
322
+ {
323
+ "epoch": 35.63,
324
+ "eval_loss": 4.0859375,
325
+ "eval_runtime": 11.7718,
326
+ "eval_samples_per_second": 68.044,
327
+ "eval_steps_per_second": 8.58,
328
+ "step": 6200
329
+ },
330
+ {
331
+ "epoch": 36.78,
332
+ "eval_loss": 4.09765625,
333
+ "eval_runtime": 11.7694,
334
+ "eval_samples_per_second": 68.058,
335
+ "eval_steps_per_second": 8.582,
336
+ "step": 6400
337
+ },
338
+ {
339
+ "epoch": 37.36,
340
+ "learning_rate": 5e-05,
341
+ "loss": 1.3947,
342
+ "step": 6500
343
+ },
344
+ {
345
+ "epoch": 37.93,
346
+ "eval_loss": 4.11328125,
347
+ "eval_runtime": 11.769,
348
+ "eval_samples_per_second": 68.06,
349
+ "eval_steps_per_second": 8.582,
350
+ "step": 6600
351
+ },
352
+ {
353
+ "epoch": 39.08,
354
+ "eval_loss": 4.1796875,
355
+ "eval_runtime": 11.7687,
356
+ "eval_samples_per_second": 68.062,
357
+ "eval_steps_per_second": 8.582,
358
+ "step": 6800
359
+ },
360
+ {
361
+ "epoch": 40.23,
362
+ "learning_rate": 5e-05,
363
+ "loss": 1.3589,
364
+ "step": 7000
365
+ },
366
+ {
367
+ "epoch": 40.23,
368
+ "eval_loss": 4.20703125,
369
+ "eval_runtime": 11.7645,
370
+ "eval_samples_per_second": 68.086,
371
+ "eval_steps_per_second": 8.585,
372
+ "step": 7000
373
+ },
374
+ {
375
+ "epoch": 41.38,
376
+ "eval_loss": 4.2734375,
377
+ "eval_runtime": 11.7699,
378
+ "eval_samples_per_second": 68.055,
379
+ "eval_steps_per_second": 8.581,
380
+ "step": 7200
381
+ },
382
+ {
383
+ "epoch": 42.53,
384
+ "eval_loss": 4.29296875,
385
+ "eval_runtime": 11.7707,
386
+ "eval_samples_per_second": 68.05,
387
+ "eval_steps_per_second": 8.581,
388
+ "step": 7400
389
+ },
390
+ {
391
+ "epoch": 43.1,
392
+ "learning_rate": 5e-05,
393
+ "loss": 1.3248,
394
+ "step": 7500
395
+ },
396
+ {
397
+ "epoch": 43.68,
398
+ "eval_loss": 4.30859375,
399
+ "eval_runtime": 11.7694,
400
+ "eval_samples_per_second": 68.058,
401
+ "eval_steps_per_second": 8.582,
402
+ "step": 7600
403
+ },
404
+ {
405
+ "epoch": 44.83,
406
+ "eval_loss": 4.34765625,
407
+ "eval_runtime": 11.7695,
408
+ "eval_samples_per_second": 68.057,
409
+ "eval_steps_per_second": 8.582,
410
+ "step": 7800
411
+ },
412
+ {
413
+ "epoch": 45.97,
414
+ "learning_rate": 5e-05,
415
+ "loss": 1.2899,
416
+ "step": 8000
417
+ },
418
+ {
419
+ "epoch": 45.97,
420
+ "eval_loss": 4.375,
421
+ "eval_runtime": 11.7703,
422
+ "eval_samples_per_second": 68.053,
423
+ "eval_steps_per_second": 8.581,
424
+ "step": 8000
425
+ },
426
+ {
427
+ "epoch": 47.13,
428
+ "eval_loss": 4.4609375,
429
+ "eval_runtime": 11.7682,
430
+ "eval_samples_per_second": 68.065,
431
+ "eval_steps_per_second": 8.582,
432
+ "step": 8200
433
+ },
434
+ {
435
+ "epoch": 48.28,
436
+ "eval_loss": 4.47265625,
437
+ "eval_runtime": 11.77,
438
+ "eval_samples_per_second": 68.054,
439
+ "eval_steps_per_second": 8.581,
440
+ "step": 8400
441
+ },
442
+ {
443
+ "epoch": 48.85,
444
+ "learning_rate": 5e-05,
445
+ "loss": 1.2585,
446
+ "step": 8500
447
+ },
448
+ {
449
+ "epoch": 49.42,
450
+ "eval_loss": 4.5546875,
451
+ "eval_runtime": 11.7706,
452
+ "eval_samples_per_second": 68.051,
453
+ "eval_steps_per_second": 8.581,
454
+ "step": 8600
455
+ },
456
+ {
457
+ "epoch": 50.57,
458
+ "eval_loss": 4.54296875,
459
+ "eval_runtime": 11.7686,
460
+ "eval_samples_per_second": 68.062,
461
+ "eval_steps_per_second": 8.582,
462
+ "step": 8800
463
+ },
464
+ {
465
+ "epoch": 51.72,
466
+ "learning_rate": 5e-05,
467
+ "loss": 1.2273,
468
+ "step": 9000
469
+ },
470
+ {
471
+ "epoch": 51.72,
472
+ "eval_loss": 4.55859375,
473
+ "eval_runtime": 11.7687,
474
+ "eval_samples_per_second": 68.062,
475
+ "eval_steps_per_second": 8.582,
476
+ "step": 9000
477
+ },
478
+ {
479
+ "epoch": 52.87,
480
+ "eval_loss": 4.58984375,
481
+ "eval_runtime": 11.7666,
482
+ "eval_samples_per_second": 68.074,
483
+ "eval_steps_per_second": 8.584,
484
+ "step": 9200
485
+ },
486
+ {
487
+ "epoch": 54.02,
488
+ "eval_loss": 4.62109375,
489
+ "eval_runtime": 11.7682,
490
+ "eval_samples_per_second": 68.065,
491
+ "eval_steps_per_second": 8.582,
492
+ "step": 9400
493
+ },
494
+ {
495
+ "epoch": 54.6,
496
+ "learning_rate": 5e-05,
497
+ "loss": 1.1995,
498
+ "step": 9500
499
+ },
500
+ {
501
+ "epoch": 55.17,
502
+ "eval_loss": 4.671875,
503
+ "eval_runtime": 11.7698,
504
+ "eval_samples_per_second": 68.056,
505
+ "eval_steps_per_second": 8.581,
506
+ "step": 9600
507
+ },
508
+ {
509
+ "epoch": 56.32,
510
+ "eval_loss": 4.71875,
511
+ "eval_runtime": 11.7677,
512
+ "eval_samples_per_second": 68.068,
513
+ "eval_steps_per_second": 8.583,
514
+ "step": 9800
515
+ },
516
+ {
517
+ "epoch": 57.47,
518
+ "learning_rate": 5e-05,
519
+ "loss": 1.1713,
520
+ "step": 10000
521
+ },
522
+ {
523
+ "epoch": 57.47,
524
+ "eval_loss": 4.72265625,
525
+ "eval_runtime": 11.7696,
526
+ "eval_samples_per_second": 68.057,
527
+ "eval_steps_per_second": 8.581,
528
+ "step": 10000
529
+ },
530
+ {
531
+ "epoch": 58.62,
532
+ "eval_loss": 4.7578125,
533
+ "eval_runtime": 11.7712,
534
+ "eval_samples_per_second": 68.047,
535
+ "eval_steps_per_second": 8.58,
536
+ "step": 10200
537
+ },
538
+ {
539
+ "epoch": 59.77,
540
+ "eval_loss": 4.77734375,
541
+ "eval_runtime": 11.7676,
542
+ "eval_samples_per_second": 68.068,
543
+ "eval_steps_per_second": 8.583,
544
+ "step": 10400
545
+ },
546
+ {
547
+ "epoch": 60.34,
548
+ "learning_rate": 5e-05,
549
+ "loss": 1.144,
550
+ "step": 10500
551
+ },
552
+ {
553
+ "epoch": 60.92,
554
+ "eval_loss": 4.78515625,
555
+ "eval_runtime": 11.7665,
556
+ "eval_samples_per_second": 68.075,
557
+ "eval_steps_per_second": 8.584,
558
+ "step": 10600
559
+ },
560
+ {
561
+ "epoch": 62.07,
562
+ "eval_loss": 4.859375,
563
+ "eval_runtime": 11.7665,
564
+ "eval_samples_per_second": 68.074,
565
+ "eval_steps_per_second": 8.584,
566
+ "step": 10800
567
+ },
568
+ {
569
+ "epoch": 63.22,
570
+ "learning_rate": 5e-05,
571
+ "loss": 1.1196,
572
+ "step": 11000
573
+ },
574
+ {
575
+ "epoch": 63.22,
576
+ "eval_loss": 4.859375,
577
+ "eval_runtime": 11.7664,
578
+ "eval_samples_per_second": 68.075,
579
+ "eval_steps_per_second": 8.584,
580
+ "step": 11000
581
+ },
582
+ {
583
+ "epoch": 64.37,
584
+ "eval_loss": 4.90234375,
585
+ "eval_runtime": 11.767,
586
+ "eval_samples_per_second": 68.072,
587
+ "eval_steps_per_second": 8.583,
588
+ "step": 11200
589
+ },
590
+ {
591
+ "epoch": 65.52,
592
+ "eval_loss": 4.921875,
593
+ "eval_runtime": 11.7663,
594
+ "eval_samples_per_second": 68.076,
595
+ "eval_steps_per_second": 8.584,
596
+ "step": 11400
597
+ },
598
+ {
599
+ "epoch": 66.09,
600
+ "learning_rate": 5e-05,
601
+ "loss": 1.0945,
602
+ "step": 11500
603
+ },
604
+ {
605
+ "epoch": 66.66,
606
+ "eval_loss": 4.94140625,
607
+ "eval_runtime": 11.7257,
608
+ "eval_samples_per_second": 68.312,
609
+ "eval_steps_per_second": 8.614,
610
+ "step": 11600
611
+ },
612
+ {
613
+ "epoch": 67.81,
614
+ "eval_loss": 4.97265625,
615
+ "eval_runtime": 11.7376,
616
+ "eval_samples_per_second": 68.242,
617
+ "eval_steps_per_second": 8.605,
618
+ "step": 11800
619
+ },
620
+ {
621
+ "epoch": 68.96,
622
+ "learning_rate": 5e-05,
623
+ "loss": 1.0698,
624
+ "step": 12000
625
+ },
626
+ {
627
+ "epoch": 68.96,
628
+ "eval_loss": 4.98046875,
629
+ "eval_runtime": 11.7405,
630
+ "eval_samples_per_second": 68.226,
631
+ "eval_steps_per_second": 8.603,
632
+ "step": 12000
633
+ },
634
+ {
635
+ "epoch": 70.11,
636
+ "eval_loss": 5.0234375,
637
+ "eval_runtime": 11.7527,
638
+ "eval_samples_per_second": 68.155,
639
+ "eval_steps_per_second": 8.594,
640
+ "step": 12200
641
+ },
642
+ {
643
+ "epoch": 71.26,
644
+ "eval_loss": 5.0546875,
645
+ "eval_runtime": 11.7518,
646
+ "eval_samples_per_second": 68.16,
647
+ "eval_steps_per_second": 8.594,
648
+ "step": 12400
649
+ },
650
+ {
651
+ "epoch": 71.84,
652
+ "learning_rate": 5e-05,
653
+ "loss": 1.047,
654
+ "step": 12500
655
+ },
656
+ {
657
+ "epoch": 72.41,
658
+ "eval_loss": 5.0859375,
659
+ "eval_runtime": 11.7522,
660
+ "eval_samples_per_second": 68.157,
661
+ "eval_steps_per_second": 8.594,
662
+ "step": 12600
663
+ },
664
+ {
665
+ "epoch": 73.56,
666
+ "eval_loss": 5.109375,
667
+ "eval_runtime": 11.7581,
668
+ "eval_samples_per_second": 68.123,
669
+ "eval_steps_per_second": 8.59,
670
+ "step": 12800
671
+ },
672
+ {
673
+ "epoch": 74.71,
674
+ "learning_rate": 5e-05,
675
+ "loss": 1.0242,
676
+ "step": 13000
677
+ },
678
+ {
679
+ "epoch": 74.71,
680
+ "eval_loss": 5.1328125,
681
+ "eval_runtime": 11.7589,
682
+ "eval_samples_per_second": 68.119,
683
+ "eval_steps_per_second": 8.589,
684
+ "step": 13000
685
+ },
686
+ {
687
+ "epoch": 75.86,
688
+ "eval_loss": 5.1484375,
689
+ "eval_runtime": 11.7654,
690
+ "eval_samples_per_second": 68.081,
691
+ "eval_steps_per_second": 8.585,
692
+ "step": 13200
693
+ },
694
+ {
695
+ "epoch": 77.01,
696
+ "eval_loss": 5.171875,
697
+ "eval_runtime": 11.7655,
698
+ "eval_samples_per_second": 68.08,
699
+ "eval_steps_per_second": 8.584,
700
+ "step": 13400
701
+ },
702
+ {
703
+ "epoch": 77.58,
704
+ "learning_rate": 5e-05,
705
+ "loss": 1.0042,
706
+ "step": 13500
707
+ },
708
+ {
709
+ "epoch": 78.16,
710
+ "eval_loss": 5.21484375,
711
+ "eval_runtime": 11.7554,
712
+ "eval_samples_per_second": 68.139,
713
+ "eval_steps_per_second": 8.592,
714
+ "step": 13600
715
+ },
716
+ {
717
+ "epoch": 79.31,
718
+ "eval_loss": 5.25390625,
719
+ "eval_runtime": 11.7677,
720
+ "eval_samples_per_second": 68.068,
721
+ "eval_steps_per_second": 8.583,
722
+ "step": 13800
723
+ },
724
+ {
725
+ "epoch": 80.46,
726
+ "learning_rate": 5e-05,
727
+ "loss": 0.983,
728
+ "step": 14000
729
+ },
730
+ {
731
+ "epoch": 80.46,
732
+ "eval_loss": 5.2421875,
733
+ "eval_runtime": 11.7633,
734
+ "eval_samples_per_second": 68.093,
735
+ "eval_steps_per_second": 8.586,
736
+ "step": 14000
737
+ },
738
+ {
739
+ "epoch": 81.61,
740
+ "eval_loss": 5.27734375,
741
+ "eval_runtime": 11.7597,
742
+ "eval_samples_per_second": 68.114,
743
+ "eval_steps_per_second": 8.589,
744
+ "step": 14200
745
+ },
746
+ {
747
+ "epoch": 82.76,
748
+ "eval_loss": 5.28515625,
749
+ "eval_runtime": 11.7676,
750
+ "eval_samples_per_second": 68.068,
751
+ "eval_steps_per_second": 8.583,
752
+ "step": 14400
753
+ },
754
+ {
755
+ "epoch": 83.33,
756
+ "learning_rate": 5e-05,
757
+ "loss": 0.9641,
758
+ "step": 14500
759
+ },
760
+ {
761
+ "epoch": 83.91,
762
+ "eval_loss": 5.32421875,
763
+ "eval_runtime": 11.7617,
764
+ "eval_samples_per_second": 68.102,
765
+ "eval_steps_per_second": 8.587,
766
+ "step": 14600
767
+ },
768
+ {
769
+ "epoch": 85.06,
770
+ "eval_loss": 5.375,
771
+ "eval_runtime": 11.7644,
772
+ "eval_samples_per_second": 68.087,
773
+ "eval_steps_per_second": 8.585,
774
+ "step": 14800
775
+ },
776
+ {
777
+ "epoch": 86.21,
778
+ "learning_rate": 5e-05,
779
+ "loss": 0.9448,
780
+ "step": 15000
781
+ },
782
+ {
783
+ "epoch": 86.21,
784
+ "eval_loss": 5.38671875,
785
+ "eval_runtime": 11.7479,
786
+ "eval_samples_per_second": 68.182,
787
+ "eval_steps_per_second": 8.597,
788
+ "step": 15000
789
+ },
790
+ {
791
+ "epoch": 87.36,
792
+ "eval_loss": 5.40625,
793
+ "eval_runtime": 11.7632,
794
+ "eval_samples_per_second": 68.094,
795
+ "eval_steps_per_second": 8.586,
796
+ "step": 15200
797
+ },
798
+ {
799
+ "epoch": 88.5,
800
+ "eval_loss": 5.41796875,
801
+ "eval_runtime": 11.7578,
802
+ "eval_samples_per_second": 68.125,
803
+ "eval_steps_per_second": 8.59,
804
+ "step": 15400
805
+ },
806
+ {
807
+ "epoch": 89.08,
808
+ "learning_rate": 5e-05,
809
+ "loss": 0.9253,
810
+ "step": 15500
811
+ },
812
+ {
813
+ "epoch": 89.65,
814
+ "eval_loss": 5.453125,
815
+ "eval_runtime": 11.7633,
816
+ "eval_samples_per_second": 68.093,
817
+ "eval_steps_per_second": 8.586,
818
+ "step": 15600
819
+ },
820
+ {
821
+ "epoch": 90.8,
822
+ "eval_loss": 5.44921875,
823
+ "eval_runtime": 11.755,
824
+ "eval_samples_per_second": 68.141,
825
+ "eval_steps_per_second": 8.592,
826
+ "step": 15800
827
+ },
828
+ {
829
+ "epoch": 91.95,
830
+ "learning_rate": 5e-05,
831
+ "loss": 0.907,
832
+ "step": 16000
833
+ },
834
+ {
835
+ "epoch": 91.95,
836
+ "eval_loss": 5.484375,
837
+ "eval_runtime": 11.7719,
838
+ "eval_samples_per_second": 68.044,
839
+ "eval_steps_per_second": 8.58,
840
+ "step": 16000
841
+ },
842
+ {
843
+ "epoch": 93.1,
844
+ "eval_loss": 5.54296875,
845
+ "eval_runtime": 11.7524,
846
+ "eval_samples_per_second": 68.157,
847
+ "eval_steps_per_second": 8.594,
848
+ "step": 16200
849
+ },
850
+ {
851
+ "epoch": 94.25,
852
+ "eval_loss": 5.56640625,
853
+ "eval_runtime": 11.762,
854
+ "eval_samples_per_second": 68.1,
855
+ "eval_steps_per_second": 8.587,
856
+ "step": 16400
857
+ },
858
+ {
859
+ "epoch": 94.83,
860
+ "learning_rate": 5e-05,
861
+ "loss": 0.8889,
862
+ "step": 16500
863
+ },
864
+ {
865
+ "epoch": 95.4,
866
+ "eval_loss": 5.58984375,
867
+ "eval_runtime": 11.7505,
868
+ "eval_samples_per_second": 68.167,
869
+ "eval_steps_per_second": 8.595,
870
+ "step": 16600
871
+ },
872
+ {
873
+ "epoch": 96.55,
874
+ "eval_loss": 5.59375,
875
+ "eval_runtime": 11.7537,
876
+ "eval_samples_per_second": 68.149,
877
+ "eval_steps_per_second": 8.593,
878
+ "step": 16800
879
+ },
880
+ {
881
+ "epoch": 97.7,
882
+ "learning_rate": 5e-05,
883
+ "loss": 0.8717,
884
+ "step": 17000
885
+ },
886
+ {
887
+ "epoch": 97.7,
888
+ "eval_loss": 5.6328125,
889
+ "eval_runtime": 11.7619,
890
+ "eval_samples_per_second": 68.101,
891
+ "eval_steps_per_second": 8.587,
892
+ "step": 17000
893
+ },
894
+ {
895
+ "epoch": 98.85,
896
+ "eval_loss": 5.64453125,
897
+ "eval_runtime": 11.7552,
898
+ "eval_samples_per_second": 68.14,
899
+ "eval_steps_per_second": 8.592,
900
+ "step": 17200
901
+ },
902
+ {
903
+ "epoch": 100.0,
904
+ "eval_loss": 5.671875,
905
+ "eval_runtime": 11.7611,
906
+ "eval_samples_per_second": 68.106,
907
+ "eval_steps_per_second": 8.588,
908
+ "step": 17400
909
+ },
910
+ {
911
+ "epoch": 100.0,
912
+ "step": 17400,
913
+ "total_flos": 2.1345852491772723e+17,
914
+ "train_loss": 0.3251776333512931,
915
+ "train_runtime": 8837.2329,
916
+ "train_samples_per_second": 31.594,
917
+ "train_steps_per_second": 1.969
918
+ }
919
+ ],
920
+ "max_steps": 17400,
921
+ "num_train_epochs": 100,
922
+ "total_flos": 2.1345852491772723e+17,
923
+ "trial_name": null,
924
+ "trial_params": null
925
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:11a054ae2c8e8206e2e83da8d39c58f0f78c45b02b9ce6323d649b9d9e8013c7
3
+ size 3375