AlekseyKorshuk commited on
Commit
5aeb850
1 Parent(s): 4a240dd

huggingartists

Browse files
README.md CHANGED
@@ -45,15 +45,15 @@ from datasets import load_dataset
45
  dataset = load_dataset("huggingartists/big-baby-tape")
46
  ```
47
 
48
- [Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/1htarx7k/artifacts), which is tracked with [W&B artifacts](https://docs.wandb.com/artifacts) at every step of the pipeline.
49
 
50
  ## Training procedure
51
 
52
  The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on Big Baby Tape's lyrics.
53
 
54
- Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/3g1oe5ka) for full transparency and reproducibility.
55
 
56
- At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/3g1oe5ka/artifacts) is logged and versioned.
57
 
58
  ## How to use
59
 
 
45
  dataset = load_dataset("huggingartists/big-baby-tape")
46
  ```
47
 
48
+ [Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/2oh33los/artifacts), which is tracked with [W&B artifacts](https://docs.wandb.com/artifacts) at every step of the pipeline.
49
 
50
  ## Training procedure
51
 
52
  The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on Big Baby Tape's lyrics.
53
 
54
+ Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/1s6lnq7u) for full transparency and reproducibility.
55
 
56
+ At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/1s6lnq7u/artifacts) is logged and versioned.
57
 
58
  ## How to use
59
 
config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "_name_or_path": "gpt2",
3
  "activation_function": "gelu_new",
4
  "architectures": [
5
  "GPT2LMHeadModel"
 
1
  {
2
+ "_name_or_path": "huggingartists/big-baby-tape",
3
  "activation_function": "gelu_new",
4
  "architectures": [
5
  "GPT2LMHeadModel"
evaluation.txt CHANGED
@@ -1 +1 @@
1
- {"eval_loss": 1.8946211338043213, "eval_runtime": 10.5235, "eval_samples_per_second": 22.141, "eval_steps_per_second": 2.851, "epoch": 1.0}
 
1
+ {"eval_loss": 1.7459986209869385, "eval_runtime": 9.6814, "eval_samples_per_second": 20.142, "eval_steps_per_second": 2.582, "epoch": 4.0}
flax_model.msgpack CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:44d87f70a283f33a59661b0fe30fe1f9cff19d9800e930e3708095a5e6458c56
3
  size 497764120
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eb5c95a97e20f7c95d543dc5984fc5037b700d9f4b39e1b5e4f6b58ce0293f31
3
  size 497764120
optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5840e665a12f250aefcbd09f8ad23143611ffa62d50914515f5b3acf3bd54c9d
3
- size 995603825
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:234a4acde292f927cb5081f20b7a098d4310346b486b808347df772453d38fbe
3
+ size 995604017
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3b476da6a207efc2b6a2780dae8d0dc6d3f8797b977d1cbb20db331183108af2
3
  size 510403817
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:97121acb26a6fa4c7fb4a741eb682f5cf8f1b3bb9f337fe691b8c152067e55fe
3
  size 510403817
rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:911921c48e2767fcdd364e55b8ac5adb7fa66719512f66f920a291ef0ff95179
3
- size 14503
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:077dc7f31533b68ea237703ffc91a5ac35fd4765522f824d8d2330befe10761d
3
+ size 14567
scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:00dd6b6c01d14ec8bb955c14a18bd11e818291c6d3cf3ac44048cedcc0186d76
3
  size 623
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bac8ceca19c5c4936fbf77b46507ea6d5819a16be52083fb75390b76c3994a44
3
  size 623
tokenizer_config.json CHANGED
@@ -1 +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", "tokenizer_class": "GPT2Tokenizer"}
 
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": "huggingartists/big-baby-tape", "tokenizer_class": "GPT2Tokenizer"}
trainer_state.json CHANGED
@@ -1,174 +1,672 @@
1
  {
2
- "best_metric": 1.8946211338043213,
3
- "best_model_checkpoint": "output/big-baby-tape/checkpoint-126",
4
- "epoch": 1.0,
5
- "global_step": 126,
6
  "is_hyper_param_search": false,
7
  "is_local_process_zero": true,
8
  "is_world_process_zero": true,
9
  "log_history": [
10
  {
11
  "epoch": 0.04,
12
- "learning_rate": 0.0001366676095727718,
13
- "loss": 2.7486,
14
  "step": 5
15
  },
16
  {
17
  "epoch": 0.08,
18
- "learning_rate": 0.00013507870183531476,
19
- "loss": 2.6211,
20
  "step": 10
21
  },
22
  {
23
  "epoch": 0.12,
24
- "learning_rate": 0.0001324579391569924,
25
- "loss": 2.2625,
26
  "step": 15
27
  },
28
  {
29
- "epoch": 0.16,
30
- "learning_rate": 0.00012884599993319768,
31
- "loss": 2.3757,
32
  "step": 20
33
  },
34
  {
35
- "epoch": 0.2,
36
- "learning_rate": 0.00012429894719210777,
37
- "loss": 2.3396,
38
  "step": 25
39
  },
40
  {
41
- "epoch": 0.24,
42
- "learning_rate": 0.00011888735840752609,
43
- "loss": 2.2627,
44
  "step": 30
45
  },
46
  {
47
- "epoch": 0.28,
48
- "learning_rate": 0.00011269523002449659,
49
- "loss": 2.226,
50
  "step": 35
51
  },
52
  {
53
- "epoch": 0.32,
54
- "learning_rate": 0.0001058186737011911,
55
- "loss": 2.1916,
56
  "step": 40
57
  },
58
  {
59
- "epoch": 0.36,
60
- "learning_rate": 9.836442450346448e-05,
61
- "loss": 2.2987,
62
  "step": 45
63
  },
64
  {
65
- "epoch": 0.4,
66
- "learning_rate": 9.044818420726556e-05,
67
- "loss": 2.0771,
68
  "step": 50
69
  },
70
  {
71
- "epoch": 0.44,
72
- "learning_rate": 8.219282542347867e-05,
73
- "loss": 2.1651,
74
  "step": 55
75
  },
76
  {
77
- "epoch": 0.48,
78
- "learning_rate": 7.372648442002871e-05,
79
- "loss": 2.1705,
80
  "step": 60
81
  },
82
  {
83
- "epoch": 0.52,
84
- "learning_rate": 6.518057224367617e-05,
85
- "loss": 2.0684,
86
  "step": 65
87
  },
88
  {
89
- "epoch": 0.56,
90
- "learning_rate": 5.668773501204858e-05,
91
- "loss": 2.025,
92
  "step": 70
93
  },
94
  {
95
- "epoch": 0.6,
96
- "learning_rate": 4.837979503541197e-05,
97
- "loss": 2.1328,
98
  "step": 75
99
  },
100
  {
101
- "epoch": 0.63,
102
- "learning_rate": 4.0385704725240065e-05,
103
- "loss": 2.076,
104
  "step": 80
105
  },
106
  {
107
- "epoch": 0.67,
108
- "learning_rate": 3.282954504816644e-05,
109
- "loss": 2.0965,
110
  "step": 85
111
  },
112
  {
113
- "epoch": 0.71,
114
- "learning_rate": 2.5828599592490882e-05,
115
- "loss": 2.157,
116
  "step": 90
117
  },
118
  {
119
- "epoch": 0.75,
120
- "learning_rate": 1.9491534140783314e-05,
121
- "loss": 2.0402,
122
  "step": 95
123
  },
124
  {
125
- "epoch": 0.79,
126
- "learning_rate": 1.3916710004507539e-05,
127
- "loss": 2.0006,
128
  "step": 100
129
  },
130
  {
131
- "epoch": 0.83,
132
- "learning_rate": 9.190657300387505e-06,
133
- "loss": 1.993,
134
  "step": 105
135
  },
136
  {
137
- "epoch": 0.87,
138
- "learning_rate": 5.38673186569003e-06,
139
- "loss": 2.0394,
140
  "step": 110
141
  },
142
  {
143
- "epoch": 0.91,
144
- "learning_rate": 2.5639766592291746e-06,
145
- "loss": 1.9176,
146
  "step": 115
147
  },
148
  {
149
- "epoch": 0.95,
150
- "learning_rate": 7.662053209561833e-07,
151
- "loss": 2.0078,
152
  "step": 120
153
  },
154
  {
155
- "epoch": 0.99,
156
- "learning_rate": 2.132211474400556e-08,
157
- "loss": 2.0426,
158
  "step": 125
159
  },
160
  {
161
  "epoch": 1.0,
162
- "eval_loss": 1.8946211338043213,
163
- "eval_runtime": 10.3833,
164
- "eval_samples_per_second": 22.44,
165
- "eval_steps_per_second": 2.889,
166
- "step": 126
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
167
  }
168
  ],
169
- "max_steps": 126,
170
- "num_train_epochs": 1,
171
- "total_flos": 130776662016000.0,
172
  "trial_name": null,
173
  "trial_params": null
174
  }
 
1
  {
2
+ "best_metric": 1.7459986209869385,
3
+ "best_model_checkpoint": "output/big-baby-tape/checkpoint-520",
4
+ "epoch": 4.0,
5
+ "global_step": 520,
6
  "is_hyper_param_search": false,
7
  "is_local_process_zero": true,
8
  "is_world_process_zero": true,
9
  "log_history": [
10
  {
11
  "epoch": 0.04,
12
+ "learning_rate": 5.001712368734975e-07,
13
+ "loss": 1.923,
14
  "step": 5
15
  },
16
  {
17
  "epoch": 0.08,
18
+ "learning_rate": 1.9933913245728244e-06,
19
+ "loss": 2.0321,
20
  "step": 10
21
  },
22
  {
23
  "epoch": 0.12,
24
+ "learning_rate": 4.457885751780542e-06,
25
+ "loss": 2.0426,
26
  "step": 15
27
  },
28
  {
29
+ "epoch": 0.15,
30
+ "learning_rate": 7.857716640189785e-06,
31
+ "loss": 2.0776,
32
  "step": 20
33
  },
34
  {
35
+ "epoch": 0.19,
36
+ "learning_rate": 1.2143306799695167e-05,
37
+ "loss": 1.9711,
38
  "step": 25
39
  },
40
  {
41
+ "epoch": 0.23,
42
+ "learning_rate": 1.7252162675462473e-05,
43
+ "loss": 2.0637,
44
  "step": 30
45
  },
46
  {
47
+ "epoch": 0.27,
48
+ "learning_rate": 2.3109785644681427e-05,
49
+ "loss": 1.8713,
50
  "step": 35
51
  },
52
  {
53
+ "epoch": 0.31,
54
+ "learning_rate": 2.9630758374242703e-05,
55
+ "loss": 2.088,
56
  "step": 40
57
  },
58
  {
59
+ "epoch": 0.35,
60
+ "learning_rate": 3.671999039779749e-05,
61
+ "loss": 1.9995,
62
  "step": 45
63
  },
64
  {
65
+ "epoch": 0.38,
66
+ "learning_rate": 4.427410474888204e-05,
67
+ "loss": 2.0004,
68
  "step": 50
69
  },
70
  {
71
+ "epoch": 0.42,
72
+ "learning_rate": 5.218294542987354e-05,
73
+ "loss": 1.9791,
74
  "step": 55
75
  },
76
  {
77
+ "epoch": 0.46,
78
+ "learning_rate": 6.03311837344848e-05,
79
+ "loss": 1.882,
80
  "step": 60
81
  },
82
  {
83
+ "epoch": 0.5,
84
+ "learning_rate": 6.859999999999999e-05,
85
+ "loss": 1.8736,
86
  "step": 65
87
  },
88
  {
89
+ "epoch": 0.54,
90
+ "learning_rate": 7.686881626551516e-05,
91
+ "loss": 1.9173,
92
  "step": 70
93
  },
94
  {
95
+ "epoch": 0.58,
96
+ "learning_rate": 8.501705457012643e-05,
97
+ "loss": 2.0122,
98
  "step": 75
99
  },
100
  {
101
+ "epoch": 0.62,
102
+ "learning_rate": 9.292589525111794e-05,
103
+ "loss": 1.871,
104
  "step": 80
105
  },
106
  {
107
+ "epoch": 0.65,
108
+ "learning_rate": 0.00010048000960220248,
109
+ "loss": 2.0619,
110
  "step": 85
111
  },
112
  {
113
+ "epoch": 0.69,
114
+ "learning_rate": 0.00010756924162575728,
115
+ "loss": 1.9238,
116
  "step": 90
117
  },
118
  {
119
+ "epoch": 0.73,
120
+ "learning_rate": 0.00011409021435531856,
121
+ "loss": 2.0238,
122
  "step": 95
123
  },
124
  {
125
+ "epoch": 0.77,
126
+ "learning_rate": 0.0001199478373245375,
127
+ "loss": 1.9745,
128
  "step": 100
129
  },
130
  {
131
+ "epoch": 0.81,
132
+ "learning_rate": 0.00012505669320030482,
133
+ "loss": 1.9809,
134
  "step": 105
135
  },
136
  {
137
+ "epoch": 0.85,
138
+ "learning_rate": 0.0001293422833598102,
139
+ "loss": 1.8881,
140
  "step": 110
141
  },
142
  {
143
+ "epoch": 0.88,
144
+ "learning_rate": 0.00013274211424821946,
145
+ "loss": 1.8788,
146
  "step": 115
147
  },
148
  {
149
+ "epoch": 0.92,
150
+ "learning_rate": 0.00013520660867542716,
151
+ "loss": 1.984,
152
  "step": 120
153
  },
154
  {
155
+ "epoch": 0.96,
156
+ "learning_rate": 0.00013669982876312649,
157
+ "loss": 1.9443,
158
  "step": 125
159
  },
160
  {
161
  "epoch": 1.0,
162
+ "learning_rate": 0.0001372,
163
+ "loss": 1.9006,
164
+ "step": 130
165
+ },
166
+ {
167
+ "epoch": 1.0,
168
+ "eval_loss": 1.9121332168579102,
169
+ "eval_runtime": 9.6453,
170
+ "eval_samples_per_second": 20.217,
171
+ "eval_steps_per_second": 2.592,
172
+ "step": 130
173
+ },
174
+ {
175
+ "epoch": 1.04,
176
+ "learning_rate": 0.0001366998287631265,
177
+ "loss": 1.8831,
178
+ "step": 135
179
+ },
180
+ {
181
+ "epoch": 1.08,
182
+ "learning_rate": 0.00013520660867542716,
183
+ "loss": 1.8547,
184
+ "step": 140
185
+ },
186
+ {
187
+ "epoch": 1.12,
188
+ "learning_rate": 0.00013274211424821946,
189
+ "loss": 1.87,
190
+ "step": 145
191
+ },
192
+ {
193
+ "epoch": 1.15,
194
+ "learning_rate": 0.00012934228335981023,
195
+ "loss": 1.8612,
196
+ "step": 150
197
+ },
198
+ {
199
+ "epoch": 1.19,
200
+ "learning_rate": 0.00012505669320030482,
201
+ "loss": 1.7875,
202
+ "step": 155
203
+ },
204
+ {
205
+ "epoch": 1.23,
206
+ "learning_rate": 0.00011994783732453754,
207
+ "loss": 1.8327,
208
+ "step": 160
209
+ },
210
+ {
211
+ "epoch": 1.27,
212
+ "learning_rate": 0.00011409021435531858,
213
+ "loss": 1.9541,
214
+ "step": 165
215
+ },
216
+ {
217
+ "epoch": 1.31,
218
+ "learning_rate": 0.00010756924162575734,
219
+ "loss": 1.8205,
220
+ "step": 170
221
+ },
222
+ {
223
+ "epoch": 1.35,
224
+ "learning_rate": 0.00010048000960220251,
225
+ "loss": 1.8993,
226
+ "step": 175
227
+ },
228
+ {
229
+ "epoch": 1.38,
230
+ "learning_rate": 9.292589525111797e-05,
231
+ "loss": 1.8431,
232
+ "step": 180
233
+ },
234
+ {
235
+ "epoch": 1.42,
236
+ "learning_rate": 8.501705457012652e-05,
237
+ "loss": 1.8127,
238
+ "step": 185
239
+ },
240
+ {
241
+ "epoch": 1.46,
242
+ "learning_rate": 7.686881626551514e-05,
243
+ "loss": 1.8428,
244
+ "step": 190
245
+ },
246
+ {
247
+ "epoch": 1.5,
248
+ "learning_rate": 6.860000000000001e-05,
249
+ "loss": 1.7509,
250
+ "step": 195
251
+ },
252
+ {
253
+ "epoch": 1.54,
254
+ "learning_rate": 6.03311837344849e-05,
255
+ "loss": 1.7859,
256
+ "step": 200
257
+ },
258
+ {
259
+ "epoch": 1.58,
260
+ "learning_rate": 5.218294542987351e-05,
261
+ "loss": 1.8358,
262
+ "step": 205
263
+ },
264
+ {
265
+ "epoch": 1.62,
266
+ "learning_rate": 4.427410474888207e-05,
267
+ "loss": 1.7393,
268
+ "step": 210
269
+ },
270
+ {
271
+ "epoch": 1.65,
272
+ "learning_rate": 3.6719990397797524e-05,
273
+ "loss": 1.8051,
274
+ "step": 215
275
+ },
276
+ {
277
+ "epoch": 1.69,
278
+ "learning_rate": 2.9630758374242683e-05,
279
+ "loss": 1.8083,
280
+ "step": 220
281
+ },
282
+ {
283
+ "epoch": 1.73,
284
+ "learning_rate": 2.310978564468145e-05,
285
+ "loss": 1.7062,
286
+ "step": 225
287
+ },
288
+ {
289
+ "epoch": 1.77,
290
+ "learning_rate": 1.7252162675462497e-05,
291
+ "loss": 1.7829,
292
+ "step": 230
293
+ },
294
+ {
295
+ "epoch": 1.81,
296
+ "learning_rate": 1.214330679969522e-05,
297
+ "loss": 1.7874,
298
+ "step": 235
299
+ },
300
+ {
301
+ "epoch": 1.85,
302
+ "learning_rate": 7.8577166401898e-06,
303
+ "loss": 1.8435,
304
+ "step": 240
305
+ },
306
+ {
307
+ "epoch": 1.88,
308
+ "learning_rate": 4.457885751780558e-06,
309
+ "loss": 1.8288,
310
+ "step": 245
311
+ },
312
+ {
313
+ "epoch": 1.92,
314
+ "learning_rate": 1.9933913245728472e-06,
315
+ "loss": 1.7984,
316
+ "step": 250
317
+ },
318
+ {
319
+ "epoch": 1.96,
320
+ "learning_rate": 5.001712368734975e-07,
321
+ "loss": 1.8331,
322
+ "step": 255
323
+ },
324
+ {
325
+ "epoch": 2.0,
326
+ "learning_rate": 0.0,
327
+ "loss": 1.8175,
328
+ "step": 260
329
+ },
330
+ {
331
+ "epoch": 2.0,
332
+ "eval_loss": 1.8163535594940186,
333
+ "eval_runtime": 9.657,
334
+ "eval_samples_per_second": 20.193,
335
+ "eval_steps_per_second": 2.589,
336
+ "step": 260
337
+ },
338
+ {
339
+ "epoch": 2.04,
340
+ "learning_rate": 5.001712368734899e-07,
341
+ "loss": 1.7297,
342
+ "step": 265
343
+ },
344
+ {
345
+ "epoch": 2.08,
346
+ "learning_rate": 1.9933913245728396e-06,
347
+ "loss": 1.5719,
348
+ "step": 270
349
+ },
350
+ {
351
+ "epoch": 2.12,
352
+ "learning_rate": 4.457885751780535e-06,
353
+ "loss": 1.7261,
354
+ "step": 275
355
+ },
356
+ {
357
+ "epoch": 2.15,
358
+ "learning_rate": 7.857716640189778e-06,
359
+ "loss": 1.6366,
360
+ "step": 280
361
+ },
362
+ {
363
+ "epoch": 2.19,
364
+ "learning_rate": 1.2143306799695189e-05,
365
+ "loss": 1.8032,
366
+ "step": 285
367
+ },
368
+ {
369
+ "epoch": 2.23,
370
+ "learning_rate": 1.725216267546246e-05,
371
+ "loss": 1.8041,
372
+ "step": 290
373
+ },
374
+ {
375
+ "epoch": 2.27,
376
+ "learning_rate": 2.310978564468141e-05,
377
+ "loss": 1.752,
378
+ "step": 295
379
+ },
380
+ {
381
+ "epoch": 2.31,
382
+ "learning_rate": 2.9630758374242642e-05,
383
+ "loss": 1.831,
384
+ "step": 300
385
+ },
386
+ {
387
+ "epoch": 2.35,
388
+ "learning_rate": 3.671999039779748e-05,
389
+ "loss": 1.6489,
390
+ "step": 305
391
+ },
392
+ {
393
+ "epoch": 2.38,
394
+ "learning_rate": 4.427410474888202e-05,
395
+ "loss": 1.7224,
396
+ "step": 310
397
+ },
398
+ {
399
+ "epoch": 2.42,
400
+ "learning_rate": 5.218294542987346e-05,
401
+ "loss": 1.6862,
402
+ "step": 315
403
+ },
404
+ {
405
+ "epoch": 2.46,
406
+ "learning_rate": 6.033118373448485e-05,
407
+ "loss": 1.7129,
408
+ "step": 320
409
+ },
410
+ {
411
+ "epoch": 2.5,
412
+ "learning_rate": 6.859999999999997e-05,
413
+ "loss": 1.9049,
414
+ "step": 325
415
+ },
416
+ {
417
+ "epoch": 2.54,
418
+ "learning_rate": 7.68688162655151e-05,
419
+ "loss": 1.8259,
420
+ "step": 330
421
+ },
422
+ {
423
+ "epoch": 2.58,
424
+ "learning_rate": 8.501705457012648e-05,
425
+ "loss": 1.7537,
426
+ "step": 335
427
+ },
428
+ {
429
+ "epoch": 2.62,
430
+ "learning_rate": 9.292589525111793e-05,
431
+ "loss": 1.7193,
432
+ "step": 340
433
+ },
434
+ {
435
+ "epoch": 2.65,
436
+ "learning_rate": 0.00010048000960220248,
437
+ "loss": 1.7921,
438
+ "step": 345
439
+ },
440
+ {
441
+ "epoch": 2.69,
442
+ "learning_rate": 0.00010756924162575731,
443
+ "loss": 1.7727,
444
+ "step": 350
445
+ },
446
+ {
447
+ "epoch": 2.73,
448
+ "learning_rate": 0.00011409021435531856,
449
+ "loss": 1.8267,
450
+ "step": 355
451
+ },
452
+ {
453
+ "epoch": 2.77,
454
+ "learning_rate": 0.0001199478373245375,
455
+ "loss": 1.6066,
456
+ "step": 360
457
+ },
458
+ {
459
+ "epoch": 2.81,
460
+ "learning_rate": 0.0001250566932003048,
461
+ "loss": 1.757,
462
+ "step": 365
463
+ },
464
+ {
465
+ "epoch": 2.85,
466
+ "learning_rate": 0.00012934228335981018,
467
+ "loss": 1.7909,
468
+ "step": 370
469
+ },
470
+ {
471
+ "epoch": 2.88,
472
+ "learning_rate": 0.00013274211424821943,
473
+ "loss": 1.7481,
474
+ "step": 375
475
+ },
476
+ {
477
+ "epoch": 2.92,
478
+ "learning_rate": 0.00013520660867542716,
479
+ "loss": 1.7786,
480
+ "step": 380
481
+ },
482
+ {
483
+ "epoch": 2.96,
484
+ "learning_rate": 0.00013669982876312649,
485
+ "loss": 1.689,
486
+ "step": 385
487
+ },
488
+ {
489
+ "epoch": 3.0,
490
+ "learning_rate": 0.0001372,
491
+ "loss": 1.7754,
492
+ "step": 390
493
+ },
494
+ {
495
+ "epoch": 3.0,
496
+ "eval_loss": 1.8065567016601562,
497
+ "eval_runtime": 9.6575,
498
+ "eval_samples_per_second": 20.192,
499
+ "eval_steps_per_second": 2.589,
500
+ "step": 390
501
+ },
502
+ {
503
+ "epoch": 3.04,
504
+ "learning_rate": 0.0001366998287631265,
505
+ "loss": 1.6829,
506
+ "step": 395
507
+ },
508
+ {
509
+ "epoch": 3.08,
510
+ "learning_rate": 0.0001352066086754272,
511
+ "loss": 1.7295,
512
+ "step": 400
513
+ },
514
+ {
515
+ "epoch": 3.12,
516
+ "learning_rate": 0.0001327421142482195,
517
+ "loss": 1.6853,
518
+ "step": 405
519
+ },
520
+ {
521
+ "epoch": 3.15,
522
+ "learning_rate": 0.00012934228335981015,
523
+ "loss": 1.7343,
524
+ "step": 410
525
+ },
526
+ {
527
+ "epoch": 3.19,
528
+ "learning_rate": 0.00012505669320030482,
529
+ "loss": 1.7685,
530
+ "step": 415
531
+ },
532
+ {
533
+ "epoch": 3.23,
534
+ "learning_rate": 0.00011994783732453755,
535
+ "loss": 1.6945,
536
+ "step": 420
537
+ },
538
+ {
539
+ "epoch": 3.27,
540
+ "learning_rate": 0.00011409021435531858,
541
+ "loss": 1.7482,
542
+ "step": 425
543
+ },
544
+ {
545
+ "epoch": 3.31,
546
+ "learning_rate": 0.00010756924162575738,
547
+ "loss": 1.5529,
548
+ "step": 430
549
+ },
550
+ {
551
+ "epoch": 3.35,
552
+ "learning_rate": 0.00010048000960220263,
553
+ "loss": 1.7141,
554
+ "step": 435
555
+ },
556
+ {
557
+ "epoch": 3.38,
558
+ "learning_rate": 9.292589525111788e-05,
559
+ "loss": 1.6804,
560
+ "step": 440
561
+ },
562
+ {
563
+ "epoch": 3.42,
564
+ "learning_rate": 8.501705457012643e-05,
565
+ "loss": 1.7525,
566
+ "step": 445
567
+ },
568
+ {
569
+ "epoch": 3.46,
570
+ "learning_rate": 7.686881626551516e-05,
571
+ "loss": 1.6325,
572
+ "step": 450
573
+ },
574
+ {
575
+ "epoch": 3.5,
576
+ "learning_rate": 6.860000000000003e-05,
577
+ "loss": 1.6941,
578
+ "step": 455
579
+ },
580
+ {
581
+ "epoch": 3.54,
582
+ "learning_rate": 6.033118373448492e-05,
583
+ "loss": 1.5753,
584
+ "step": 460
585
+ },
586
+ {
587
+ "epoch": 3.58,
588
+ "learning_rate": 5.218294542987365e-05,
589
+ "loss": 1.6565,
590
+ "step": 465
591
+ },
592
+ {
593
+ "epoch": 3.62,
594
+ "learning_rate": 4.42741047488822e-05,
595
+ "loss": 1.6553,
596
+ "step": 470
597
+ },
598
+ {
599
+ "epoch": 3.65,
600
+ "learning_rate": 3.671999039779743e-05,
601
+ "loss": 1.5705,
602
+ "step": 475
603
+ },
604
+ {
605
+ "epoch": 3.69,
606
+ "learning_rate": 2.9630758374242696e-05,
607
+ "loss": 1.7043,
608
+ "step": 480
609
+ },
610
+ {
611
+ "epoch": 3.73,
612
+ "learning_rate": 2.3109785644681465e-05,
613
+ "loss": 1.5999,
614
+ "step": 485
615
+ },
616
+ {
617
+ "epoch": 3.77,
618
+ "learning_rate": 1.7252162675462504e-05,
619
+ "loss": 1.675,
620
+ "step": 490
621
+ },
622
+ {
623
+ "epoch": 3.81,
624
+ "learning_rate": 1.2143306799695228e-05,
625
+ "loss": 1.6187,
626
+ "step": 495
627
+ },
628
+ {
629
+ "epoch": 3.85,
630
+ "learning_rate": 7.857716640189861e-06,
631
+ "loss": 1.5483,
632
+ "step": 500
633
+ },
634
+ {
635
+ "epoch": 3.88,
636
+ "learning_rate": 4.4578857517805195e-06,
637
+ "loss": 1.6583,
638
+ "step": 505
639
+ },
640
+ {
641
+ "epoch": 3.92,
642
+ "learning_rate": 1.9933913245728244e-06,
643
+ "loss": 1.6295,
644
+ "step": 510
645
+ },
646
+ {
647
+ "epoch": 3.96,
648
+ "learning_rate": 5.001712368734975e-07,
649
+ "loss": 1.5502,
650
+ "step": 515
651
+ },
652
+ {
653
+ "epoch": 4.0,
654
+ "learning_rate": 0.0,
655
+ "loss": 1.7657,
656
+ "step": 520
657
+ },
658
+ {
659
+ "epoch": 4.0,
660
+ "eval_loss": 1.7459986209869385,
661
+ "eval_runtime": 9.6571,
662
+ "eval_samples_per_second": 20.192,
663
+ "eval_steps_per_second": 2.589,
664
+ "step": 520
665
  }
666
  ],
667
+ "max_steps": 520,
668
+ "num_train_epochs": 4,
669
+ "total_flos": 542964842496000.0,
670
  "trial_name": null,
671
  "trial_params": null
672
  }
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a697d9fe12925fe593db615b6532711275101c30a75091651a57d54878a90dd9
3
  size 2671
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5b5708c7447842968121331f651a0667504a3cf69e6dddb04ddfb6f7d888825b
3
  size 2671