weizhepei commited on
Commit
688237c
·
verified ·
1 Parent(s): fda5164

Model save

Browse files
Files changed (5) hide show
  1. README.md +58 -0
  2. all_results.json +8 -0
  3. generation_config.json +12 -0
  4. train_results.json +8 -0
  5. trainer_state.json +1281 -0
README.md ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: meta-llama/Llama-3.1-8B-Instruct
3
+ library_name: transformers
4
+ model_name: Llama3.1-8B-WebArena-Lite-SFT-CoT-QwQ-32B-epoch-3
5
+ tags:
6
+ - generated_from_trainer
7
+ - trl
8
+ - sft
9
+ licence: license
10
+ ---
11
+
12
+ # Model Card for Llama3.1-8B-WebArena-Lite-SFT-CoT-QwQ-32B-epoch-3
13
+
14
+ This model is a fine-tuned version of [meta-llama/Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct).
15
+ It has been trained using [TRL](https://github.com/huggingface/trl).
16
+
17
+ ## Quick start
18
+
19
+ ```python
20
+ from transformers import pipeline
21
+
22
+ question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
23
+ generator = pipeline("text-generation", model="weizhepei/Llama3.1-8B-WebArena-Lite-SFT-CoT-QwQ-32B-epoch-3", device="cuda")
24
+ output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
25
+ print(output["generated_text"])
26
+ ```
27
+
28
+ ## Training procedure
29
+
30
+ [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/uva-llm/huggingface/runs/fwb20lnf)
31
+
32
+
33
+ This model was trained with SFT.
34
+
35
+ ### Framework versions
36
+
37
+ - TRL: 0.16.0.dev0
38
+ - Transformers: 4.49.0
39
+ - Pytorch: 2.5.1
40
+ - Datasets: 3.4.1
41
+ - Tokenizers: 0.21.1
42
+
43
+ ## Citations
44
+
45
+
46
+
47
+ Cite TRL as:
48
+
49
+ ```bibtex
50
+ @misc{vonwerra2022trl,
51
+ title = {{TRL: Transformer Reinforcement Learning}},
52
+ author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
53
+ year = 2020,
54
+ journal = {GitHub repository},
55
+ publisher = {GitHub},
56
+ howpublished = {\url{https://github.com/huggingface/trl}}
57
+ }
58
+ ```
all_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "total_flos": 412579921461248.0,
3
+ "train_loss": 0.5179592104175607,
4
+ "train_runtime": 13894.0345,
5
+ "train_samples": 9460,
6
+ "train_samples_per_second": 2.043,
7
+ "train_steps_per_second": 0.064
8
+ }
generation_config.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 128000,
3
+ "do_sample": true,
4
+ "eos_token_id": [
5
+ 128001,
6
+ 128008,
7
+ 128009
8
+ ],
9
+ "temperature": 0.6,
10
+ "top_p": 0.9,
11
+ "transformers_version": "4.49.0"
12
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "total_flos": 412579921461248.0,
3
+ "train_loss": 0.5179592104175607,
4
+ "train_runtime": 13894.0345,
5
+ "train_samples": 9460,
6
+ "train_samples_per_second": 2.043,
7
+ "train_steps_per_second": 0.064
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,1281 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 3.0,
5
+ "eval_steps": 500,
6
+ "global_step": 888,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.016891891891891893,
13
+ "grad_norm": 6.231681321080043,
14
+ "learning_rate": 5.555555555555556e-06,
15
+ "loss": 1.2707,
16
+ "step": 5
17
+ },
18
+ {
19
+ "epoch": 0.033783783783783786,
20
+ "grad_norm": 4.950872417286887,
21
+ "learning_rate": 1.1111111111111112e-05,
22
+ "loss": 1.1132,
23
+ "step": 10
24
+ },
25
+ {
26
+ "epoch": 0.05067567567567568,
27
+ "grad_norm": 3.967491399410514,
28
+ "learning_rate": 1.6666666666666667e-05,
29
+ "loss": 0.9841,
30
+ "step": 15
31
+ },
32
+ {
33
+ "epoch": 0.06756756756756757,
34
+ "grad_norm": 3.3129251745959687,
35
+ "learning_rate": 2.2222222222222223e-05,
36
+ "loss": 0.9505,
37
+ "step": 20
38
+ },
39
+ {
40
+ "epoch": 0.08445945945945946,
41
+ "grad_norm": 3.2010737153707294,
42
+ "learning_rate": 2.777777777777778e-05,
43
+ "loss": 0.9554,
44
+ "step": 25
45
+ },
46
+ {
47
+ "epoch": 0.10135135135135136,
48
+ "grad_norm": 2.9508900364586625,
49
+ "learning_rate": 3.3333333333333335e-05,
50
+ "loss": 0.9444,
51
+ "step": 30
52
+ },
53
+ {
54
+ "epoch": 0.11824324324324324,
55
+ "grad_norm": 2.8841214572197207,
56
+ "learning_rate": 3.888888888888889e-05,
57
+ "loss": 0.9727,
58
+ "step": 35
59
+ },
60
+ {
61
+ "epoch": 0.13513513513513514,
62
+ "grad_norm": 2.6403601369490213,
63
+ "learning_rate": 4.4444444444444447e-05,
64
+ "loss": 0.9681,
65
+ "step": 40
66
+ },
67
+ {
68
+ "epoch": 0.15202702702702703,
69
+ "grad_norm": 2.71201222699466,
70
+ "learning_rate": 5e-05,
71
+ "loss": 1.0178,
72
+ "step": 45
73
+ },
74
+ {
75
+ "epoch": 0.16891891891891891,
76
+ "grad_norm": 2.4481643042636048,
77
+ "learning_rate": 4.999609406904966e-05,
78
+ "loss": 1.0114,
79
+ "step": 50
80
+ },
81
+ {
82
+ "epoch": 0.1858108108108108,
83
+ "grad_norm": 2.2579319180534676,
84
+ "learning_rate": 4.99843776323139e-05,
85
+ "loss": 0.9717,
86
+ "step": 55
87
+ },
88
+ {
89
+ "epoch": 0.20270270270270271,
90
+ "grad_norm": 2.6055124227126965,
91
+ "learning_rate": 4.9964854757667645e-05,
92
+ "loss": 0.9995,
93
+ "step": 60
94
+ },
95
+ {
96
+ "epoch": 0.2195945945945946,
97
+ "grad_norm": 2.2286636459152467,
98
+ "learning_rate": 4.993753222333314e-05,
99
+ "loss": 0.9817,
100
+ "step": 65
101
+ },
102
+ {
103
+ "epoch": 0.23648648648648649,
104
+ "grad_norm": 2.6802583991959867,
105
+ "learning_rate": 4.9902419515526604e-05,
106
+ "loss": 0.9936,
107
+ "step": 70
108
+ },
109
+ {
110
+ "epoch": 0.2533783783783784,
111
+ "grad_norm": 2.2015258103078423,
112
+ "learning_rate": 4.985952882516468e-05,
113
+ "loss": 0.9776,
114
+ "step": 75
115
+ },
116
+ {
117
+ "epoch": 0.2702702702702703,
118
+ "grad_norm": 1.9745512455187957,
119
+ "learning_rate": 4.980887504363181e-05,
120
+ "loss": 0.9709,
121
+ "step": 80
122
+ },
123
+ {
124
+ "epoch": 0.28716216216216217,
125
+ "grad_norm": 2.691281356514746,
126
+ "learning_rate": 4.975047575761004e-05,
127
+ "loss": 0.9705,
128
+ "step": 85
129
+ },
130
+ {
131
+ "epoch": 0.30405405405405406,
132
+ "grad_norm": 1.8744903969564979,
133
+ "learning_rate": 4.9684351242973035e-05,
134
+ "loss": 0.9381,
135
+ "step": 90
136
+ },
137
+ {
138
+ "epoch": 0.32094594594594594,
139
+ "grad_norm": 2.499819434542776,
140
+ "learning_rate": 4.961052445774642e-05,
141
+ "loss": 0.9408,
142
+ "step": 95
143
+ },
144
+ {
145
+ "epoch": 0.33783783783783783,
146
+ "grad_norm": 2.0213491561872856,
147
+ "learning_rate": 4.952902103413691e-05,
148
+ "loss": 0.9576,
149
+ "step": 100
150
+ },
151
+ {
152
+ "epoch": 0.3547297297297297,
153
+ "grad_norm": 2.007777582220033,
154
+ "learning_rate": 4.943986926963291e-05,
155
+ "loss": 0.9538,
156
+ "step": 105
157
+ },
158
+ {
159
+ "epoch": 0.3716216216216216,
160
+ "grad_norm": 1.8376742795926042,
161
+ "learning_rate": 4.934310011717988e-05,
162
+ "loss": 0.9588,
163
+ "step": 110
164
+ },
165
+ {
166
+ "epoch": 0.3885135135135135,
167
+ "grad_norm": 2.1090401102371423,
168
+ "learning_rate": 4.923874717443361e-05,
169
+ "loss": 0.9777,
170
+ "step": 115
171
+ },
172
+ {
173
+ "epoch": 0.40540540540540543,
174
+ "grad_norm": 1.9167602630034615,
175
+ "learning_rate": 4.9126846672095325e-05,
176
+ "loss": 0.9162,
177
+ "step": 120
178
+ },
179
+ {
180
+ "epoch": 0.4222972972972973,
181
+ "grad_norm": 2.0456990567414644,
182
+ "learning_rate": 4.900743746133262e-05,
183
+ "loss": 0.9386,
184
+ "step": 125
185
+ },
186
+ {
187
+ "epoch": 0.4391891891891892,
188
+ "grad_norm": 1.9096406870985225,
189
+ "learning_rate": 4.888056100029056e-05,
190
+ "loss": 0.9624,
191
+ "step": 130
192
+ },
193
+ {
194
+ "epoch": 0.4560810810810811,
195
+ "grad_norm": 1.9809440749049774,
196
+ "learning_rate": 4.87462613396977e-05,
197
+ "loss": 0.9257,
198
+ "step": 135
199
+ },
200
+ {
201
+ "epoch": 0.47297297297297297,
202
+ "grad_norm": 1.7258775836343794,
203
+ "learning_rate": 4.860458510757189e-05,
204
+ "loss": 0.9254,
205
+ "step": 140
206
+ },
207
+ {
208
+ "epoch": 0.48986486486486486,
209
+ "grad_norm": 1.9425639247796285,
210
+ "learning_rate": 4.845558149303135e-05,
211
+ "loss": 0.9476,
212
+ "step": 145
213
+ },
214
+ {
215
+ "epoch": 0.5067567567567568,
216
+ "grad_norm": 1.7602777895956392,
217
+ "learning_rate": 4.8299302229216506e-05,
218
+ "loss": 0.9615,
219
+ "step": 150
220
+ },
221
+ {
222
+ "epoch": 0.5236486486486487,
223
+ "grad_norm": 1.6992288952271115,
224
+ "learning_rate": 4.8135801575328546e-05,
225
+ "loss": 0.8942,
226
+ "step": 155
227
+ },
228
+ {
229
+ "epoch": 0.5405405405405406,
230
+ "grad_norm": 1.698637849222003,
231
+ "learning_rate": 4.796513629779098e-05,
232
+ "loss": 0.8944,
233
+ "step": 160
234
+ },
235
+ {
236
+ "epoch": 0.5574324324324325,
237
+ "grad_norm": 1.5467487223203435,
238
+ "learning_rate": 4.7787365650540663e-05,
239
+ "loss": 0.9153,
240
+ "step": 165
241
+ },
242
+ {
243
+ "epoch": 0.5743243243243243,
244
+ "grad_norm": 1.6334496828072615,
245
+ "learning_rate": 4.760255135445521e-05,
246
+ "loss": 0.9248,
247
+ "step": 170
248
+ },
249
+ {
250
+ "epoch": 0.5912162162162162,
251
+ "grad_norm": 1.6259508720656166,
252
+ "learning_rate": 4.741075757592389e-05,
253
+ "loss": 0.8847,
254
+ "step": 175
255
+ },
256
+ {
257
+ "epoch": 0.6081081081081081,
258
+ "grad_norm": 1.790410553077543,
259
+ "learning_rate": 4.721205090456941e-05,
260
+ "loss": 0.9289,
261
+ "step": 180
262
+ },
263
+ {
264
+ "epoch": 0.625,
265
+ "grad_norm": 1.8532634400660235,
266
+ "learning_rate": 4.7006500330128473e-05,
267
+ "loss": 0.9015,
268
+ "step": 185
269
+ },
270
+ {
271
+ "epoch": 0.6418918918918919,
272
+ "grad_norm": 1.7875929509109747,
273
+ "learning_rate": 4.679417721849889e-05,
274
+ "loss": 0.8905,
275
+ "step": 190
276
+ },
277
+ {
278
+ "epoch": 0.6587837837837838,
279
+ "grad_norm": 1.5944547606006951,
280
+ "learning_rate": 4.657515528696183e-05,
281
+ "loss": 0.8929,
282
+ "step": 195
283
+ },
284
+ {
285
+ "epoch": 0.6756756756756757,
286
+ "grad_norm": 1.5660749975109045,
287
+ "learning_rate": 4.6349510578587635e-05,
288
+ "loss": 0.8757,
289
+ "step": 200
290
+ },
291
+ {
292
+ "epoch": 0.6925675675675675,
293
+ "grad_norm": 1.6423933991915096,
294
+ "learning_rate": 4.61173214358341e-05,
295
+ "loss": 0.8934,
296
+ "step": 205
297
+ },
298
+ {
299
+ "epoch": 0.7094594594594594,
300
+ "grad_norm": 1.559933642024552,
301
+ "learning_rate": 4.5878668473346464e-05,
302
+ "loss": 0.9102,
303
+ "step": 210
304
+ },
305
+ {
306
+ "epoch": 0.7263513513513513,
307
+ "grad_norm": 1.5955677484161037,
308
+ "learning_rate": 4.5633634549968514e-05,
309
+ "loss": 0.8886,
310
+ "step": 215
311
+ },
312
+ {
313
+ "epoch": 0.7432432432432432,
314
+ "grad_norm": 1.5548188228815374,
315
+ "learning_rate": 4.538230473997449e-05,
316
+ "loss": 0.8737,
317
+ "step": 220
318
+ },
319
+ {
320
+ "epoch": 0.7601351351351351,
321
+ "grad_norm": 1.6430486936390023,
322
+ "learning_rate": 4.512476630353183e-05,
323
+ "loss": 0.8651,
324
+ "step": 225
325
+ },
326
+ {
327
+ "epoch": 0.777027027027027,
328
+ "grad_norm": 1.497167470194,
329
+ "learning_rate": 4.4861108656404955e-05,
330
+ "loss": 0.8724,
331
+ "step": 230
332
+ },
333
+ {
334
+ "epoch": 0.793918918918919,
335
+ "grad_norm": 1.4259126734193541,
336
+ "learning_rate": 4.459142333891067e-05,
337
+ "loss": 0.8905,
338
+ "step": 235
339
+ },
340
+ {
341
+ "epoch": 0.8108108108108109,
342
+ "grad_norm": 1.4831105867234027,
343
+ "learning_rate": 4.4315803984135965e-05,
344
+ "loss": 0.8619,
345
+ "step": 240
346
+ },
347
+ {
348
+ "epoch": 0.8277027027027027,
349
+ "grad_norm": 1.5825985585703166,
350
+ "learning_rate": 4.403434628542914e-05,
351
+ "loss": 0.8619,
352
+ "step": 245
353
+ },
354
+ {
355
+ "epoch": 0.8445945945945946,
356
+ "grad_norm": 1.5675917133492445,
357
+ "learning_rate": 4.374714796317566e-05,
358
+ "loss": 0.9152,
359
+ "step": 250
360
+ },
361
+ {
362
+ "epoch": 0.8614864864864865,
363
+ "grad_norm": 1.5361276988078942,
364
+ "learning_rate": 4.345430873087026e-05,
365
+ "loss": 0.8688,
366
+ "step": 255
367
+ },
368
+ {
369
+ "epoch": 0.8783783783783784,
370
+ "grad_norm": 1.369553191214301,
371
+ "learning_rate": 4.315593026049703e-05,
372
+ "loss": 0.851,
373
+ "step": 260
374
+ },
375
+ {
376
+ "epoch": 0.8952702702702703,
377
+ "grad_norm": 1.6638541563918563,
378
+ "learning_rate": 4.28521161472295e-05,
379
+ "loss": 0.8734,
380
+ "step": 265
381
+ },
382
+ {
383
+ "epoch": 0.9121621621621622,
384
+ "grad_norm": 1.441713025001977,
385
+ "learning_rate": 4.2542971873463075e-05,
386
+ "loss": 0.8433,
387
+ "step": 270
388
+ },
389
+ {
390
+ "epoch": 0.9290540540540541,
391
+ "grad_norm": 1.4960913750522002,
392
+ "learning_rate": 4.222860477219215e-05,
393
+ "loss": 0.8442,
394
+ "step": 275
395
+ },
396
+ {
397
+ "epoch": 0.9459459459459459,
398
+ "grad_norm": 1.462307887191494,
399
+ "learning_rate": 4.190912398974478e-05,
400
+ "loss": 0.8515,
401
+ "step": 280
402
+ },
403
+ {
404
+ "epoch": 0.9628378378378378,
405
+ "grad_norm": 1.3528940388287598,
406
+ "learning_rate": 4.158464044788774e-05,
407
+ "loss": 0.8438,
408
+ "step": 285
409
+ },
410
+ {
411
+ "epoch": 0.9797297297297297,
412
+ "grad_norm": 1.4502619959473744,
413
+ "learning_rate": 4.125526680531517e-05,
414
+ "loss": 0.8417,
415
+ "step": 290
416
+ },
417
+ {
418
+ "epoch": 0.9966216216216216,
419
+ "grad_norm": 1.3664408170375748,
420
+ "learning_rate": 4.092111741853414e-05,
421
+ "loss": 0.8355,
422
+ "step": 295
423
+ },
424
+ {
425
+ "epoch": 1.0135135135135136,
426
+ "grad_norm": 1.48215782409135,
427
+ "learning_rate": 4.058230830216084e-05,
428
+ "loss": 0.605,
429
+ "step": 300
430
+ },
431
+ {
432
+ "epoch": 1.0304054054054055,
433
+ "grad_norm": 1.325024180793296,
434
+ "learning_rate": 4.0238957088640935e-05,
435
+ "loss": 0.5206,
436
+ "step": 305
437
+ },
438
+ {
439
+ "epoch": 1.0472972972972974,
440
+ "grad_norm": 1.5992825466313396,
441
+ "learning_rate": 3.989118298740837e-05,
442
+ "loss": 0.5089,
443
+ "step": 310
444
+ },
445
+ {
446
+ "epoch": 1.0641891891891893,
447
+ "grad_norm": 1.4273213416805155,
448
+ "learning_rate": 3.953910674349652e-05,
449
+ "loss": 0.4936,
450
+ "step": 315
451
+ },
452
+ {
453
+ "epoch": 1.0810810810810811,
454
+ "grad_norm": 1.3287056923860654,
455
+ "learning_rate": 3.9182850595616334e-05,
456
+ "loss": 0.4983,
457
+ "step": 320
458
+ },
459
+ {
460
+ "epoch": 1.097972972972973,
461
+ "grad_norm": 1.4134836690592798,
462
+ "learning_rate": 3.882253823371574e-05,
463
+ "loss": 0.4897,
464
+ "step": 325
465
+ },
466
+ {
467
+ "epoch": 1.114864864864865,
468
+ "grad_norm": 1.302428194474921,
469
+ "learning_rate": 3.8458294756035284e-05,
470
+ "loss": 0.511,
471
+ "step": 330
472
+ },
473
+ {
474
+ "epoch": 1.1317567567567568,
475
+ "grad_norm": 1.4644457082188613,
476
+ "learning_rate": 3.809024662567478e-05,
477
+ "loss": 0.5208,
478
+ "step": 335
479
+ },
480
+ {
481
+ "epoch": 1.1486486486486487,
482
+ "grad_norm": 1.7300512446405285,
483
+ "learning_rate": 3.771852162668611e-05,
484
+ "loss": 0.5015,
485
+ "step": 340
486
+ },
487
+ {
488
+ "epoch": 1.1655405405405406,
489
+ "grad_norm": 1.467044194490883,
490
+ "learning_rate": 3.734324881970736e-05,
491
+ "loss": 0.5272,
492
+ "step": 345
493
+ },
494
+ {
495
+ "epoch": 1.1824324324324325,
496
+ "grad_norm": 1.2979658306396822,
497
+ "learning_rate": 3.69645584971538e-05,
498
+ "loss": 0.5032,
499
+ "step": 350
500
+ },
501
+ {
502
+ "epoch": 1.1993243243243243,
503
+ "grad_norm": 1.3933445860127016,
504
+ "learning_rate": 3.658258213798108e-05,
505
+ "loss": 0.4999,
506
+ "step": 355
507
+ },
508
+ {
509
+ "epoch": 1.2162162162162162,
510
+ "grad_norm": 1.3589175020707676,
511
+ "learning_rate": 3.619745236203666e-05,
512
+ "loss": 0.5183,
513
+ "step": 360
514
+ },
515
+ {
516
+ "epoch": 1.2331081081081081,
517
+ "grad_norm": 1.4074778914667276,
518
+ "learning_rate": 3.580930288401491e-05,
519
+ "loss": 0.5073,
520
+ "step": 365
521
+ },
522
+ {
523
+ "epoch": 1.25,
524
+ "grad_norm": 1.3086984542811302,
525
+ "learning_rate": 3.541826846703224e-05,
526
+ "loss": 0.4942,
527
+ "step": 370
528
+ },
529
+ {
530
+ "epoch": 1.2668918918918919,
531
+ "grad_norm": 1.4381479516212257,
532
+ "learning_rate": 3.5024484875838145e-05,
533
+ "loss": 0.5071,
534
+ "step": 375
535
+ },
536
+ {
537
+ "epoch": 1.2837837837837838,
538
+ "grad_norm": 1.2976783376198013,
539
+ "learning_rate": 3.462808882967856e-05,
540
+ "loss": 0.5278,
541
+ "step": 380
542
+ },
543
+ {
544
+ "epoch": 1.3006756756756757,
545
+ "grad_norm": 1.3793720906248872,
546
+ "learning_rate": 3.4229217954827716e-05,
547
+ "loss": 0.5166,
548
+ "step": 385
549
+ },
550
+ {
551
+ "epoch": 1.3175675675675675,
552
+ "grad_norm": 1.3862580674531486,
553
+ "learning_rate": 3.3828010736805185e-05,
554
+ "loss": 0.5077,
555
+ "step": 390
556
+ },
557
+ {
558
+ "epoch": 1.3344594594594594,
559
+ "grad_norm": 1.3562667437179257,
560
+ "learning_rate": 3.342460647229459e-05,
561
+ "loss": 0.5048,
562
+ "step": 395
563
+ },
564
+ {
565
+ "epoch": 1.3513513513513513,
566
+ "grad_norm": 1.374243227751759,
567
+ "learning_rate": 3.301914522078055e-05,
568
+ "loss": 0.4957,
569
+ "step": 400
570
+ },
571
+ {
572
+ "epoch": 1.3682432432432432,
573
+ "grad_norm": 1.3910923941785058,
574
+ "learning_rate": 3.261176775592097e-05,
575
+ "loss": 0.4864,
576
+ "step": 405
577
+ },
578
+ {
579
+ "epoch": 1.385135135135135,
580
+ "grad_norm": 1.4190426820456954,
581
+ "learning_rate": 3.220261551667128e-05,
582
+ "loss": 0.4908,
583
+ "step": 410
584
+ },
585
+ {
586
+ "epoch": 1.402027027027027,
587
+ "grad_norm": 1.3067413904671548,
588
+ "learning_rate": 3.179183055817767e-05,
589
+ "loss": 0.4997,
590
+ "step": 415
591
+ },
592
+ {
593
+ "epoch": 1.4189189189189189,
594
+ "grad_norm": 1.4464186039630254,
595
+ "learning_rate": 3.137955550245643e-05,
596
+ "loss": 0.4888,
597
+ "step": 420
598
+ },
599
+ {
600
+ "epoch": 1.4358108108108107,
601
+ "grad_norm": 1.3944491264799062,
602
+ "learning_rate": 3.096593348887647e-05,
603
+ "loss": 0.5026,
604
+ "step": 425
605
+ },
606
+ {
607
+ "epoch": 1.4527027027027026,
608
+ "grad_norm": 1.2998412690334573,
609
+ "learning_rate": 3.055110812446221e-05,
610
+ "loss": 0.4755,
611
+ "step": 430
612
+ },
613
+ {
614
+ "epoch": 1.4695945945945945,
615
+ "grad_norm": 1.277085043268365,
616
+ "learning_rate": 3.0135223434034053e-05,
617
+ "loss": 0.4703,
618
+ "step": 435
619
+ },
620
+ {
621
+ "epoch": 1.4864864864864864,
622
+ "grad_norm": 1.2464902265161508,
623
+ "learning_rate": 2.9718423810203944e-05,
624
+ "loss": 0.5012,
625
+ "step": 440
626
+ },
627
+ {
628
+ "epoch": 1.5033783783783785,
629
+ "grad_norm": 1.289453338271534,
630
+ "learning_rate": 2.9300853963243056e-05,
631
+ "loss": 0.5004,
632
+ "step": 445
633
+ },
634
+ {
635
+ "epoch": 1.5202702702702702,
636
+ "grad_norm": 1.2772690540474154,
637
+ "learning_rate": 2.8882658870839326e-05,
638
+ "loss": 0.493,
639
+ "step": 450
640
+ },
641
+ {
642
+ "epoch": 1.5371621621621623,
643
+ "grad_norm": 1.286041286434996,
644
+ "learning_rate": 2.8463983727762046e-05,
645
+ "loss": 0.5165,
646
+ "step": 455
647
+ },
648
+ {
649
+ "epoch": 1.554054054054054,
650
+ "grad_norm": 1.3335486353524793,
651
+ "learning_rate": 2.8044973895451166e-05,
652
+ "loss": 0.4804,
653
+ "step": 460
654
+ },
655
+ {
656
+ "epoch": 1.570945945945946,
657
+ "grad_norm": 1.3917515334345343,
658
+ "learning_rate": 2.7625774851548708e-05,
659
+ "loss": 0.4764,
660
+ "step": 465
661
+ },
662
+ {
663
+ "epoch": 1.5878378378378377,
664
+ "grad_norm": 1.3172805254924786,
665
+ "learning_rate": 2.7206532139389767e-05,
666
+ "loss": 0.4754,
667
+ "step": 470
668
+ },
669
+ {
670
+ "epoch": 1.6047297297297298,
671
+ "grad_norm": 1.3246558687025178,
672
+ "learning_rate": 2.67873913174708e-05,
673
+ "loss": 0.4858,
674
+ "step": 475
675
+ },
676
+ {
677
+ "epoch": 1.6216216216216215,
678
+ "grad_norm": 1.3328679629033688,
679
+ "learning_rate": 2.6368497908912592e-05,
680
+ "loss": 0.5026,
681
+ "step": 480
682
+ },
683
+ {
684
+ "epoch": 1.6385135135135136,
685
+ "grad_norm": 1.343169827036629,
686
+ "learning_rate": 2.5949997350935534e-05,
687
+ "loss": 0.487,
688
+ "step": 485
689
+ },
690
+ {
691
+ "epoch": 1.6554054054054053,
692
+ "grad_norm": 1.2317833751218523,
693
+ "learning_rate": 2.5532034944364712e-05,
694
+ "loss": 0.4796,
695
+ "step": 490
696
+ },
697
+ {
698
+ "epoch": 1.6722972972972974,
699
+ "grad_norm": 1.25207138650032,
700
+ "learning_rate": 2.511475580318233e-05,
701
+ "loss": 0.4685,
702
+ "step": 495
703
+ },
704
+ {
705
+ "epoch": 1.689189189189189,
706
+ "grad_norm": 1.2916401389153942,
707
+ "learning_rate": 2.4698304804145038e-05,
708
+ "loss": 0.4732,
709
+ "step": 500
710
+ },
711
+ {
712
+ "epoch": 1.7060810810810811,
713
+ "grad_norm": 1.309320360265987,
714
+ "learning_rate": 2.428282653648367e-05,
715
+ "loss": 0.4758,
716
+ "step": 505
717
+ },
718
+ {
719
+ "epoch": 1.722972972972973,
720
+ "grad_norm": 1.2227129579363303,
721
+ "learning_rate": 2.386846525170263e-05,
722
+ "loss": 0.4752,
723
+ "step": 510
724
+ },
725
+ {
726
+ "epoch": 1.739864864864865,
727
+ "grad_norm": 1.3338028513859421,
728
+ "learning_rate": 2.3455364813496755e-05,
729
+ "loss": 0.4705,
730
+ "step": 515
731
+ },
732
+ {
733
+ "epoch": 1.7567567567567568,
734
+ "grad_norm": 1.2501373969082323,
735
+ "learning_rate": 2.3043668647802687e-05,
736
+ "loss": 0.4633,
737
+ "step": 520
738
+ },
739
+ {
740
+ "epoch": 1.7736486486486487,
741
+ "grad_norm": 1.2830950543814998,
742
+ "learning_rate": 2.2633519693002247e-05,
743
+ "loss": 0.4889,
744
+ "step": 525
745
+ },
746
+ {
747
+ "epoch": 1.7905405405405406,
748
+ "grad_norm": 1.2838795731857355,
749
+ "learning_rate": 2.2225060350295184e-05,
750
+ "loss": 0.4634,
751
+ "step": 530
752
+ },
753
+ {
754
+ "epoch": 1.8074324324324325,
755
+ "grad_norm": 1.2938941133398314,
756
+ "learning_rate": 2.181843243425824e-05,
757
+ "loss": 0.4639,
758
+ "step": 535
759
+ },
760
+ {
761
+ "epoch": 1.8243243243243243,
762
+ "grad_norm": 1.3650137810330223,
763
+ "learning_rate": 2.1413777123608103e-05,
764
+ "loss": 0.4643,
765
+ "step": 540
766
+ },
767
+ {
768
+ "epoch": 1.8412162162162162,
769
+ "grad_norm": 1.3447974317742337,
770
+ "learning_rate": 2.1011234912184942e-05,
771
+ "loss": 0.4533,
772
+ "step": 545
773
+ },
774
+ {
775
+ "epoch": 1.8581081081081081,
776
+ "grad_norm": 1.31773148023275,
777
+ "learning_rate": 2.0610945560173865e-05,
778
+ "loss": 0.4624,
779
+ "step": 550
780
+ },
781
+ {
782
+ "epoch": 1.875,
783
+ "grad_norm": 1.2818032468001688,
784
+ "learning_rate": 2.0213048045581014e-05,
785
+ "loss": 0.4455,
786
+ "step": 555
787
+ },
788
+ {
789
+ "epoch": 1.8918918918918919,
790
+ "grad_norm": 1.3353447408158947,
791
+ "learning_rate": 1.9817680515981274e-05,
792
+ "loss": 0.4638,
793
+ "step": 560
794
+ },
795
+ {
796
+ "epoch": 1.9087837837837838,
797
+ "grad_norm": 1.3297365957452907,
798
+ "learning_rate": 1.9424980240554236e-05,
799
+ "loss": 0.4716,
800
+ "step": 565
801
+ },
802
+ {
803
+ "epoch": 1.9256756756756757,
804
+ "grad_norm": 1.1934548733440697,
805
+ "learning_rate": 1.903508356242525e-05,
806
+ "loss": 0.441,
807
+ "step": 570
808
+ },
809
+ {
810
+ "epoch": 1.9425675675675675,
811
+ "grad_norm": 1.2911193562484828,
812
+ "learning_rate": 1.864812585132787e-05,
813
+ "loss": 0.4341,
814
+ "step": 575
815
+ },
816
+ {
817
+ "epoch": 1.9594594594594594,
818
+ "grad_norm": 1.284850775623703,
819
+ "learning_rate": 1.8264241456604324e-05,
820
+ "loss": 0.448,
821
+ "step": 580
822
+ },
823
+ {
824
+ "epoch": 1.9763513513513513,
825
+ "grad_norm": 1.2349841173785179,
826
+ "learning_rate": 1.78835636605603e-05,
827
+ "loss": 0.4423,
828
+ "step": 585
829
+ },
830
+ {
831
+ "epoch": 1.9932432432432432,
832
+ "grad_norm": 1.2146186214149826,
833
+ "learning_rate": 1.7506224632190065e-05,
834
+ "loss": 0.4409,
835
+ "step": 590
836
+ },
837
+ {
838
+ "epoch": 2.010135135135135,
839
+ "grad_norm": 1.0086827563539977,
840
+ "learning_rate": 1.7132355381288255e-05,
841
+ "loss": 0.2777,
842
+ "step": 595
843
+ },
844
+ {
845
+ "epoch": 2.027027027027027,
846
+ "grad_norm": 1.342303752922279,
847
+ "learning_rate": 1.676208571296408e-05,
848
+ "loss": 0.1598,
849
+ "step": 600
850
+ },
851
+ {
852
+ "epoch": 2.043918918918919,
853
+ "grad_norm": 1.113941499846872,
854
+ "learning_rate": 1.6395544182573753e-05,
855
+ "loss": 0.1611,
856
+ "step": 605
857
+ },
858
+ {
859
+ "epoch": 2.060810810810811,
860
+ "grad_norm": 1.019799238243946,
861
+ "learning_rate": 1.603285805108689e-05,
862
+ "loss": 0.1604,
863
+ "step": 610
864
+ },
865
+ {
866
+ "epoch": 2.0777027027027026,
867
+ "grad_norm": 1.1412888682882492,
868
+ "learning_rate": 1.5674153240902257e-05,
869
+ "loss": 0.1554,
870
+ "step": 615
871
+ },
872
+ {
873
+ "epoch": 2.0945945945945947,
874
+ "grad_norm": 1.1509735977910904,
875
+ "learning_rate": 1.5319554292128323e-05,
876
+ "loss": 0.148,
877
+ "step": 620
878
+ },
879
+ {
880
+ "epoch": 2.1114864864864864,
881
+ "grad_norm": 1.0040685111146865,
882
+ "learning_rate": 1.4969184319343654e-05,
883
+ "loss": 0.1515,
884
+ "step": 625
885
+ },
886
+ {
887
+ "epoch": 2.1283783783783785,
888
+ "grad_norm": 1.1136156062268279,
889
+ "learning_rate": 1.4623164968852327e-05,
890
+ "loss": 0.1552,
891
+ "step": 630
892
+ },
893
+ {
894
+ "epoch": 2.14527027027027,
895
+ "grad_norm": 1.0777999627533803,
896
+ "learning_rate": 1.4281616376449048e-05,
897
+ "loss": 0.1562,
898
+ "step": 635
899
+ },
900
+ {
901
+ "epoch": 2.1621621621621623,
902
+ "grad_norm": 1.0504060905612465,
903
+ "learning_rate": 1.394465712570876e-05,
904
+ "loss": 0.1522,
905
+ "step": 640
906
+ },
907
+ {
908
+ "epoch": 2.179054054054054,
909
+ "grad_norm": 1.0905056621961537,
910
+ "learning_rate": 1.3612404206815144e-05,
911
+ "loss": 0.1473,
912
+ "step": 645
913
+ },
914
+ {
915
+ "epoch": 2.195945945945946,
916
+ "grad_norm": 1.19009398992958,
917
+ "learning_rate": 1.3284972975942367e-05,
918
+ "loss": 0.1516,
919
+ "step": 650
920
+ },
921
+ {
922
+ "epoch": 2.2128378378378377,
923
+ "grad_norm": 1.1365455436625145,
924
+ "learning_rate": 1.2962477115204094e-05,
925
+ "loss": 0.1497,
926
+ "step": 655
927
+ },
928
+ {
929
+ "epoch": 2.22972972972973,
930
+ "grad_norm": 1.1587962328316135,
931
+ "learning_rate": 1.2645028593183763e-05,
932
+ "loss": 0.1422,
933
+ "step": 660
934
+ },
935
+ {
936
+ "epoch": 2.2466216216216215,
937
+ "grad_norm": 1.0337185281275136,
938
+ "learning_rate": 1.233273762605982e-05,
939
+ "loss": 0.1473,
940
+ "step": 665
941
+ },
942
+ {
943
+ "epoch": 2.2635135135135136,
944
+ "grad_norm": 1.1195019763510503,
945
+ "learning_rate": 1.2025712639339268e-05,
946
+ "loss": 0.1529,
947
+ "step": 670
948
+ },
949
+ {
950
+ "epoch": 2.2804054054054053,
951
+ "grad_norm": 1.0396864482228654,
952
+ "learning_rate": 1.1724060230213075e-05,
953
+ "loss": 0.1502,
954
+ "step": 675
955
+ },
956
+ {
957
+ "epoch": 2.2972972972972974,
958
+ "grad_norm": 1.077156584604291,
959
+ "learning_rate": 1.1427885130546207e-05,
960
+ "loss": 0.1491,
961
+ "step": 680
962
+ },
963
+ {
964
+ "epoch": 2.314189189189189,
965
+ "grad_norm": 1.0424824768401733,
966
+ "learning_rate": 1.1137290170515429e-05,
967
+ "loss": 0.1442,
968
+ "step": 685
969
+ },
970
+ {
971
+ "epoch": 2.331081081081081,
972
+ "grad_norm": 1.007784936402244,
973
+ "learning_rate": 1.085237624290726e-05,
974
+ "loss": 0.1476,
975
+ "step": 690
976
+ },
977
+ {
978
+ "epoch": 2.347972972972973,
979
+ "grad_norm": 1.0916964057550667,
980
+ "learning_rate": 1.0573242268088629e-05,
981
+ "loss": 0.1394,
982
+ "step": 695
983
+ },
984
+ {
985
+ "epoch": 2.364864864864865,
986
+ "grad_norm": 1.0392881227066229,
987
+ "learning_rate": 1.0299985159662348e-05,
988
+ "loss": 0.1418,
989
+ "step": 700
990
+ },
991
+ {
992
+ "epoch": 2.3817567567567566,
993
+ "grad_norm": 1.0610168893707066,
994
+ "learning_rate": 1.0032699790819288e-05,
995
+ "loss": 0.1486,
996
+ "step": 705
997
+ },
998
+ {
999
+ "epoch": 2.3986486486486487,
1000
+ "grad_norm": 1.016875434548862,
1001
+ "learning_rate": 9.77147896139897e-06,
1002
+ "loss": 0.1358,
1003
+ "step": 710
1004
+ },
1005
+ {
1006
+ "epoch": 2.4155405405405403,
1007
+ "grad_norm": 1.111072857598883,
1008
+ "learning_rate": 9.516413365670063e-06,
1009
+ "loss": 0.1527,
1010
+ "step": 715
1011
+ },
1012
+ {
1013
+ "epoch": 2.4324324324324325,
1014
+ "grad_norm": 1.0166047165710437,
1015
+ "learning_rate": 9.267591560841876e-06,
1016
+ "loss": 0.1432,
1017
+ "step": 720
1018
+ },
1019
+ {
1020
+ "epoch": 2.4493243243243246,
1021
+ "grad_norm": 1.0288128507258207,
1022
+ "learning_rate": 9.02509993631784e-06,
1023
+ "loss": 0.1381,
1024
+ "step": 725
1025
+ },
1026
+ {
1027
+ "epoch": 2.4662162162162162,
1028
+ "grad_norm": 1.067559654789781,
1029
+ "learning_rate": 8.789022683701629e-06,
1030
+ "loss": 0.144,
1031
+ "step": 730
1032
+ },
1033
+ {
1034
+ "epoch": 2.483108108108108,
1035
+ "grad_norm": 1.0329686983765223,
1036
+ "learning_rate": 8.559441767566378e-06,
1037
+ "loss": 0.143,
1038
+ "step": 735
1039
+ },
1040
+ {
1041
+ "epoch": 2.5,
1042
+ "grad_norm": 1.002825285922701,
1043
+ "learning_rate": 8.336436896997063e-06,
1044
+ "loss": 0.1297,
1045
+ "step": 740
1046
+ },
1047
+ {
1048
+ "epoch": 2.516891891891892,
1049
+ "grad_norm": 1.0924011480602662,
1050
+ "learning_rate": 8.120085497915995e-06,
1051
+ "loss": 0.139,
1052
+ "step": 745
1053
+ },
1054
+ {
1055
+ "epoch": 2.5337837837837838,
1056
+ "grad_norm": 1.09246056663418,
1057
+ "learning_rate": 7.91046268620102e-06,
1058
+ "loss": 0.1411,
1059
+ "step": 750
1060
+ },
1061
+ {
1062
+ "epoch": 2.5506756756756754,
1063
+ "grad_norm": 1.0967904913726172,
1064
+ "learning_rate": 7.70764124160576e-06,
1065
+ "loss": 0.1375,
1066
+ "step": 755
1067
+ },
1068
+ {
1069
+ "epoch": 2.5675675675675675,
1070
+ "grad_norm": 0.9863979550244457,
1071
+ "learning_rate": 7.5116915824908985e-06,
1072
+ "loss": 0.1337,
1073
+ "step": 760
1074
+ },
1075
+ {
1076
+ "epoch": 2.5844594594594597,
1077
+ "grad_norm": 1.0802259488346817,
1078
+ "learning_rate": 7.322681741375405e-06,
1079
+ "loss": 0.1353,
1080
+ "step": 765
1081
+ },
1082
+ {
1083
+ "epoch": 2.6013513513513513,
1084
+ "grad_norm": 1.1488305384571615,
1085
+ "learning_rate": 7.140677341316013e-06,
1086
+ "loss": 0.1392,
1087
+ "step": 770
1088
+ },
1089
+ {
1090
+ "epoch": 2.618243243243243,
1091
+ "grad_norm": 1.0654092076468062,
1092
+ "learning_rate": 6.96574157312333e-06,
1093
+ "loss": 0.1289,
1094
+ "step": 775
1095
+ },
1096
+ {
1097
+ "epoch": 2.635135135135135,
1098
+ "grad_norm": 1.0616589088316162,
1099
+ "learning_rate": 6.797935173422361e-06,
1100
+ "loss": 0.1296,
1101
+ "step": 780
1102
+ },
1103
+ {
1104
+ "epoch": 2.652027027027027,
1105
+ "grad_norm": 1.0808128072586296,
1106
+ "learning_rate": 6.6373164035651246e-06,
1107
+ "loss": 0.134,
1108
+ "step": 785
1109
+ },
1110
+ {
1111
+ "epoch": 2.668918918918919,
1112
+ "grad_norm": 1.0229579739214567,
1113
+ "learning_rate": 6.483941029402675e-06,
1114
+ "loss": 0.1297,
1115
+ "step": 790
1116
+ },
1117
+ {
1118
+ "epoch": 2.685810810810811,
1119
+ "grad_norm": 1.0341349547813405,
1120
+ "learning_rate": 6.337862301923552e-06,
1121
+ "loss": 0.129,
1122
+ "step": 795
1123
+ },
1124
+ {
1125
+ "epoch": 2.7027027027027026,
1126
+ "grad_norm": 0.9953533379030133,
1127
+ "learning_rate": 6.199130938765337e-06,
1128
+ "loss": 0.1292,
1129
+ "step": 800
1130
+ },
1131
+ {
1132
+ "epoch": 2.7195945945945947,
1133
+ "grad_norm": 0.9916711500948964,
1134
+ "learning_rate": 6.067795106605817e-06,
1135
+ "loss": 0.123,
1136
+ "step": 805
1137
+ },
1138
+ {
1139
+ "epoch": 2.7364864864864864,
1140
+ "grad_norm": 1.1347113646519535,
1141
+ "learning_rate": 5.943900404439816e-06,
1142
+ "loss": 0.1245,
1143
+ "step": 810
1144
+ },
1145
+ {
1146
+ "epoch": 2.7533783783783785,
1147
+ "grad_norm": 1.0854345457051158,
1148
+ "learning_rate": 5.827489847747492e-06,
1149
+ "loss": 0.1236,
1150
+ "step": 815
1151
+ },
1152
+ {
1153
+ "epoch": 2.77027027027027,
1154
+ "grad_norm": 1.0669578176856562,
1155
+ "learning_rate": 5.718603853559626e-06,
1156
+ "loss": 0.1289,
1157
+ "step": 820
1158
+ },
1159
+ {
1160
+ "epoch": 2.7871621621621623,
1161
+ "grad_norm": 1.0718894223631414,
1162
+ "learning_rate": 5.617280226425088e-06,
1163
+ "loss": 0.124,
1164
+ "step": 825
1165
+ },
1166
+ {
1167
+ "epoch": 2.804054054054054,
1168
+ "grad_norm": 1.0280712526002755,
1169
+ "learning_rate": 5.5235541452853204e-06,
1170
+ "loss": 0.1245,
1171
+ "step": 830
1172
+ },
1173
+ {
1174
+ "epoch": 2.820945945945946,
1175
+ "grad_norm": 1.0735185449716749,
1176
+ "learning_rate": 5.437458151260425e-06,
1177
+ "loss": 0.1258,
1178
+ "step": 835
1179
+ },
1180
+ {
1181
+ "epoch": 2.8378378378378377,
1182
+ "grad_norm": 0.9948777016385901,
1183
+ "learning_rate": 5.3590221363510965e-06,
1184
+ "loss": 0.1203,
1185
+ "step": 840
1186
+ },
1187
+ {
1188
+ "epoch": 2.85472972972973,
1189
+ "grad_norm": 1.0444692285037418,
1190
+ "learning_rate": 5.288273333060287e-06,
1191
+ "loss": 0.1279,
1192
+ "step": 845
1193
+ },
1194
+ {
1195
+ "epoch": 2.8716216216216215,
1196
+ "grad_norm": 1.0593788199176337,
1197
+ "learning_rate": 5.2252363049382626e-06,
1198
+ "loss": 0.1203,
1199
+ "step": 850
1200
+ },
1201
+ {
1202
+ "epoch": 2.8885135135135136,
1203
+ "grad_norm": 1.0224848591397384,
1204
+ "learning_rate": 5.169932938054281e-06,
1205
+ "loss": 0.1242,
1206
+ "step": 855
1207
+ },
1208
+ {
1209
+ "epoch": 2.9054054054054053,
1210
+ "grad_norm": 1.0109605507100674,
1211
+ "learning_rate": 5.122382433397887e-06,
1212
+ "loss": 0.1189,
1213
+ "step": 860
1214
+ },
1215
+ {
1216
+ "epoch": 2.9222972972972974,
1217
+ "grad_norm": 1.024940692386505,
1218
+ "learning_rate": 5.082601300212445e-06,
1219
+ "loss": 0.122,
1220
+ "step": 865
1221
+ },
1222
+ {
1223
+ "epoch": 2.939189189189189,
1224
+ "grad_norm": 1.1348408537056638,
1225
+ "learning_rate": 5.050603350263229e-06,
1226
+ "loss": 0.1272,
1227
+ "step": 870
1228
+ },
1229
+ {
1230
+ "epoch": 2.956081081081081,
1231
+ "grad_norm": 1.0587237780098981,
1232
+ "learning_rate": 5.026399693042064e-06,
1233
+ "loss": 0.122,
1234
+ "step": 875
1235
+ },
1236
+ {
1237
+ "epoch": 2.972972972972973,
1238
+ "grad_norm": 1.0526311563986077,
1239
+ "learning_rate": 5.009998731910186e-06,
1240
+ "loss": 0.1239,
1241
+ "step": 880
1242
+ },
1243
+ {
1244
+ "epoch": 2.989864864864865,
1245
+ "grad_norm": 1.079605600798441,
1246
+ "learning_rate": 5.001406161180633e-06,
1247
+ "loss": 0.1251,
1248
+ "step": 885
1249
+ },
1250
+ {
1251
+ "epoch": 3.0,
1252
+ "step": 888,
1253
+ "total_flos": 412579921461248.0,
1254
+ "train_loss": 0.5179592104175607,
1255
+ "train_runtime": 13894.0345,
1256
+ "train_samples_per_second": 2.043,
1257
+ "train_steps_per_second": 0.064
1258
+ }
1259
+ ],
1260
+ "logging_steps": 5,
1261
+ "max_steps": 888,
1262
+ "num_input_tokens_seen": 0,
1263
+ "num_train_epochs": 3,
1264
+ "save_steps": 500,
1265
+ "stateful_callbacks": {
1266
+ "TrainerControl": {
1267
+ "args": {
1268
+ "should_epoch_stop": false,
1269
+ "should_evaluate": false,
1270
+ "should_log": false,
1271
+ "should_save": true,
1272
+ "should_training_stop": true
1273
+ },
1274
+ "attributes": {}
1275
+ }
1276
+ },
1277
+ "total_flos": 412579921461248.0,
1278
+ "train_batch_size": 4,
1279
+ "trial_name": null,
1280
+ "trial_params": null
1281
+ }