DrNicefellow commited on
Commit
a0219d2
1 Parent(s): f6a07f2

Upload 13 files

Browse files
config.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_function": "gelu_new",
3
+ "architectures": [
4
+ "GPT2LMHeadModel"
5
+ ],
6
+ "attn_pdrop": 0.1,
7
+ "bos_token_id": 50256,
8
+ "embd_pdrop": 0.1,
9
+ "eos_token_id": 50256,
10
+ "initializer_range": 0.02,
11
+ "layer_norm_epsilon": 1e-05,
12
+ "model_type": "gpt2",
13
+ "n_ctx": 512,
14
+ "n_embd": 1280,
15
+ "n_head": 20,
16
+ "n_inner": null,
17
+ "n_layer": 36,
18
+ "n_positions": 512,
19
+ "reorder_and_upcast_attn": false,
20
+ "resid_pdrop": 0.1,
21
+ "scale_attn_by_inverse_layer_idx": false,
22
+ "scale_attn_weights": true,
23
+ "summary_activation": null,
24
+ "summary_first_dropout": 0.1,
25
+ "summary_proj_to_labels": true,
26
+ "summary_type": "cls_index",
27
+ "summary_use_proj": true,
28
+ "task_specific_params": {
29
+ "text-generation": {
30
+ "do_sample": true,
31
+ "max_length": 50
32
+ }
33
+ },
34
+ "torch_dtype": "float32",
35
+ "transformers_version": "4.38.0.dev0",
36
+ "use_cache": true,
37
+ "vocab_size": 50257
38
+ }
generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 50256,
4
+ "eos_token_id": 50256,
5
+ "transformers_version": "4.38.0.dev0"
6
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:df31987f3fd32b3652c12822918c32ee9b11700aeda6c527458b374e42a34e87
3
+ size 6187371110
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:affe535355f1376b5fd1a5cf3b3594cb18d5c6b8c59808c2da1772f15eafc672
3
+ size 3093641690
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b9b396444c21bebf007957b13b3f4771de2c0754b2f43b937d086bbb5c27b1a4
3
+ size 14244
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:455dd772bf6f726b3416cd7289b4d980e876008e83d509033c9a1631f3031bac
3
+ size 1064
special_tokens_map.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<|endoftext|>",
3
+ "eos_token": "<|endoftext|>",
4
+ "unk_token": "<|endoftext|>"
5
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "50256": {
5
+ "content": "<|endoftext|>",
6
+ "lstrip": false,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ }
12
+ },
13
+ "bos_token": "<|endoftext|>",
14
+ "clean_up_tokenization_spaces": true,
15
+ "eos_token": "<|endoftext|>",
16
+ "model_max_length": 1024,
17
+ "tokenizer_class": "GPT2Tokenizer",
18
+ "unk_token": "<|endoftext|>"
19
+ }
trainer_state.json ADDED
@@ -0,0 +1,1401 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.0687871403531273,
5
+ "eval_steps": 100000000,
6
+ "global_step": 115000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0,
13
+ "learning_rate": 5e-05,
14
+ "loss": 7.7296,
15
+ "step": 500
16
+ },
17
+ {
18
+ "epoch": 0.0,
19
+ "learning_rate": 4.9999535305620204e-05,
20
+ "loss": 6.5786,
21
+ "step": 1000
22
+ },
23
+ {
24
+ "epoch": 0.0,
25
+ "learning_rate": 4.99990706112404e-05,
26
+ "loss": 6.2317,
27
+ "step": 1500
28
+ },
29
+ {
30
+ "epoch": 0.0,
31
+ "learning_rate": 4.99986059168606e-05,
32
+ "loss": 6.0053,
33
+ "step": 2000
34
+ },
35
+ {
36
+ "epoch": 0.0,
37
+ "learning_rate": 4.99981412224808e-05,
38
+ "loss": 5.8615,
39
+ "step": 2500
40
+ },
41
+ {
42
+ "epoch": 0.0,
43
+ "learning_rate": 4.9997676528101e-05,
44
+ "loss": 5.6794,
45
+ "step": 3000
46
+ },
47
+ {
48
+ "epoch": 0.0,
49
+ "learning_rate": 4.99972118337212e-05,
50
+ "loss": 5.5481,
51
+ "step": 3500
52
+ },
53
+ {
54
+ "epoch": 0.0,
55
+ "learning_rate": 4.9996747139341396e-05,
56
+ "loss": 5.4375,
57
+ "step": 4000
58
+ },
59
+ {
60
+ "epoch": 0.0,
61
+ "learning_rate": 4.99962824449616e-05,
62
+ "loss": 5.3391,
63
+ "step": 4500
64
+ },
65
+ {
66
+ "epoch": 0.0,
67
+ "learning_rate": 4.99958177505818e-05,
68
+ "loss": 5.2598,
69
+ "step": 5000
70
+ },
71
+ {
72
+ "epoch": 0.01,
73
+ "learning_rate": 4.9995353056202e-05,
74
+ "loss": 5.17,
75
+ "step": 5500
76
+ },
77
+ {
78
+ "epoch": 0.01,
79
+ "learning_rate": 4.9994888361822204e-05,
80
+ "loss": 5.0886,
81
+ "step": 6000
82
+ },
83
+ {
84
+ "epoch": 0.01,
85
+ "learning_rate": 4.99944236674424e-05,
86
+ "loss": 5.0128,
87
+ "step": 6500
88
+ },
89
+ {
90
+ "epoch": 0.01,
91
+ "learning_rate": 4.99939589730626e-05,
92
+ "loss": 4.9504,
93
+ "step": 7000
94
+ },
95
+ {
96
+ "epoch": 0.01,
97
+ "learning_rate": 4.99934942786828e-05,
98
+ "loss": 4.8755,
99
+ "step": 7500
100
+ },
101
+ {
102
+ "epoch": 0.01,
103
+ "learning_rate": 4.9993029584303e-05,
104
+ "loss": 4.8402,
105
+ "step": 8000
106
+ },
107
+ {
108
+ "epoch": 0.01,
109
+ "learning_rate": 4.99925648899232e-05,
110
+ "loss": 4.7674,
111
+ "step": 8500
112
+ },
113
+ {
114
+ "epoch": 0.01,
115
+ "learning_rate": 4.9992100195543396e-05,
116
+ "loss": 4.7254,
117
+ "step": 9000
118
+ },
119
+ {
120
+ "epoch": 0.01,
121
+ "learning_rate": 4.99916355011636e-05,
122
+ "loss": 4.6749,
123
+ "step": 9500
124
+ },
125
+ {
126
+ "epoch": 0.01,
127
+ "learning_rate": 4.99911708067838e-05,
128
+ "loss": 4.6135,
129
+ "step": 10000
130
+ },
131
+ {
132
+ "epoch": 0.01,
133
+ "learning_rate": 4.9990706112403995e-05,
134
+ "loss": 4.5239,
135
+ "step": 10500
136
+ },
137
+ {
138
+ "epoch": 0.01,
139
+ "learning_rate": 4.99902414180242e-05,
140
+ "loss": 4.4729,
141
+ "step": 11000
142
+ },
143
+ {
144
+ "epoch": 0.01,
145
+ "learning_rate": 4.998977672364439e-05,
146
+ "loss": 4.4434,
147
+ "step": 11500
148
+ },
149
+ {
150
+ "epoch": 0.01,
151
+ "learning_rate": 4.9989312029264594e-05,
152
+ "loss": 4.3587,
153
+ "step": 12000
154
+ },
155
+ {
156
+ "epoch": 0.01,
157
+ "learning_rate": 4.9988847334884796e-05,
158
+ "loss": 4.3126,
159
+ "step": 12500
160
+ },
161
+ {
162
+ "epoch": 0.01,
163
+ "learning_rate": 4.998838264050499e-05,
164
+ "loss": 4.2658,
165
+ "step": 13000
166
+ },
167
+ {
168
+ "epoch": 0.01,
169
+ "learning_rate": 4.998791794612519e-05,
170
+ "loss": 4.2412,
171
+ "step": 13500
172
+ },
173
+ {
174
+ "epoch": 0.01,
175
+ "learning_rate": 4.9987453251745395e-05,
176
+ "loss": 4.2035,
177
+ "step": 14000
178
+ },
179
+ {
180
+ "epoch": 0.01,
181
+ "learning_rate": 4.998698855736559e-05,
182
+ "loss": 4.1695,
183
+ "step": 14500
184
+ },
185
+ {
186
+ "epoch": 0.01,
187
+ "learning_rate": 4.998652386298579e-05,
188
+ "loss": 4.1406,
189
+ "step": 15000
190
+ },
191
+ {
192
+ "epoch": 0.01,
193
+ "learning_rate": 4.9986059168605994e-05,
194
+ "loss": 4.1275,
195
+ "step": 15500
196
+ },
197
+ {
198
+ "epoch": 0.01,
199
+ "learning_rate": 4.9985594474226196e-05,
200
+ "loss": 4.1032,
201
+ "step": 16000
202
+ },
203
+ {
204
+ "epoch": 0.02,
205
+ "learning_rate": 4.99851297798464e-05,
206
+ "loss": 4.0588,
207
+ "step": 16500
208
+ },
209
+ {
210
+ "epoch": 0.02,
211
+ "learning_rate": 4.9984665085466594e-05,
212
+ "loss": 4.0536,
213
+ "step": 17000
214
+ },
215
+ {
216
+ "epoch": 0.02,
217
+ "learning_rate": 4.9984200391086796e-05,
218
+ "loss": 3.9979,
219
+ "step": 17500
220
+ },
221
+ {
222
+ "epoch": 0.02,
223
+ "learning_rate": 4.998373569670699e-05,
224
+ "loss": 3.9913,
225
+ "step": 18000
226
+ },
227
+ {
228
+ "epoch": 0.02,
229
+ "learning_rate": 4.998327100232719e-05,
230
+ "loss": 3.988,
231
+ "step": 18500
232
+ },
233
+ {
234
+ "epoch": 0.02,
235
+ "learning_rate": 4.9982806307947395e-05,
236
+ "loss": 3.9575,
237
+ "step": 19000
238
+ },
239
+ {
240
+ "epoch": 0.02,
241
+ "learning_rate": 4.998234161356759e-05,
242
+ "loss": 3.942,
243
+ "step": 19500
244
+ },
245
+ {
246
+ "epoch": 0.02,
247
+ "learning_rate": 4.998187691918779e-05,
248
+ "loss": 3.9258,
249
+ "step": 20000
250
+ },
251
+ {
252
+ "epoch": 0.02,
253
+ "learning_rate": 4.998141222480799e-05,
254
+ "loss": 3.8736,
255
+ "step": 20500
256
+ },
257
+ {
258
+ "epoch": 0.02,
259
+ "learning_rate": 4.998094753042819e-05,
260
+ "loss": 3.903,
261
+ "step": 21000
262
+ },
263
+ {
264
+ "epoch": 0.2,
265
+ "learning_rate": 4.9804810945458605e-05,
266
+ "loss": 3.7108,
267
+ "step": 21500
268
+ },
269
+ {
270
+ "epoch": 0.2,
271
+ "learning_rate": 4.980016358701714e-05,
272
+ "loss": 3.6276,
273
+ "step": 22000
274
+ },
275
+ {
276
+ "epoch": 0.21,
277
+ "learning_rate": 4.979551622857568e-05,
278
+ "loss": 3.568,
279
+ "step": 22500
280
+ },
281
+ {
282
+ "epoch": 0.21,
283
+ "learning_rate": 4.9790868870134216e-05,
284
+ "loss": 3.5307,
285
+ "step": 23000
286
+ },
287
+ {
288
+ "epoch": 0.22,
289
+ "learning_rate": 4.978622151169275e-05,
290
+ "loss": 3.4857,
291
+ "step": 23500
292
+ },
293
+ {
294
+ "epoch": 0.22,
295
+ "learning_rate": 4.978157415325129e-05,
296
+ "loss": 3.4463,
297
+ "step": 24000
298
+ },
299
+ {
300
+ "epoch": 0.23,
301
+ "learning_rate": 4.977692679480983e-05,
302
+ "loss": 3.4396,
303
+ "step": 24500
304
+ },
305
+ {
306
+ "epoch": 0.23,
307
+ "learning_rate": 4.977227943636837e-05,
308
+ "loss": 3.402,
309
+ "step": 25000
310
+ },
311
+ {
312
+ "epoch": 0.24,
313
+ "learning_rate": 4.9767632077926904e-05,
314
+ "loss": 3.3716,
315
+ "step": 25500
316
+ },
317
+ {
318
+ "epoch": 0.24,
319
+ "learning_rate": 4.9762984719485446e-05,
320
+ "loss": 3.3533,
321
+ "step": 26000
322
+ },
323
+ {
324
+ "epoch": 0.25,
325
+ "learning_rate": 4.975833736104398e-05,
326
+ "loss": 3.3084,
327
+ "step": 26500
328
+ },
329
+ {
330
+ "epoch": 0.25,
331
+ "learning_rate": 4.975369000260252e-05,
332
+ "loss": 3.293,
333
+ "step": 27000
334
+ },
335
+ {
336
+ "epoch": 0.26,
337
+ "learning_rate": 4.974904264416106e-05,
338
+ "loss": 3.2707,
339
+ "step": 27500
340
+ },
341
+ {
342
+ "epoch": 0.26,
343
+ "learning_rate": 4.97443952857196e-05,
344
+ "loss": 3.2561,
345
+ "step": 28000
346
+ },
347
+ {
348
+ "epoch": 0.26,
349
+ "learning_rate": 4.9739747927278134e-05,
350
+ "loss": 3.2219,
351
+ "step": 28500
352
+ },
353
+ {
354
+ "epoch": 0.27,
355
+ "learning_rate": 4.9735100568836676e-05,
356
+ "loss": 3.2014,
357
+ "step": 29000
358
+ },
359
+ {
360
+ "epoch": 0.27,
361
+ "learning_rate": 4.973045321039521e-05,
362
+ "loss": 3.1892,
363
+ "step": 29500
364
+ },
365
+ {
366
+ "epoch": 0.28,
367
+ "learning_rate": 4.972580585195375e-05,
368
+ "loss": 3.1734,
369
+ "step": 30000
370
+ },
371
+ {
372
+ "epoch": 0.28,
373
+ "learning_rate": 4.972115849351229e-05,
374
+ "loss": 3.1624,
375
+ "step": 30500
376
+ },
377
+ {
378
+ "epoch": 0.29,
379
+ "learning_rate": 4.971651113507082e-05,
380
+ "loss": 3.1416,
381
+ "step": 31000
382
+ },
383
+ {
384
+ "epoch": 0.29,
385
+ "learning_rate": 4.9711863776629364e-05,
386
+ "loss": 3.1255,
387
+ "step": 31500
388
+ },
389
+ {
390
+ "epoch": 0.3,
391
+ "learning_rate": 4.97072164181879e-05,
392
+ "loss": 3.1074,
393
+ "step": 32000
394
+ },
395
+ {
396
+ "epoch": 0.3,
397
+ "learning_rate": 4.970256905974644e-05,
398
+ "loss": 3.0992,
399
+ "step": 32500
400
+ },
401
+ {
402
+ "epoch": 0.31,
403
+ "learning_rate": 4.9697921701304976e-05,
404
+ "loss": 3.0867,
405
+ "step": 33000
406
+ },
407
+ {
408
+ "epoch": 0.31,
409
+ "learning_rate": 4.969327434286352e-05,
410
+ "loss": 3.0711,
411
+ "step": 33500
412
+ },
413
+ {
414
+ "epoch": 0.32,
415
+ "learning_rate": 4.968862698442205e-05,
416
+ "loss": 3.0621,
417
+ "step": 34000
418
+ },
419
+ {
420
+ "epoch": 0.32,
421
+ "learning_rate": 4.9683979625980594e-05,
422
+ "loss": 3.0543,
423
+ "step": 34500
424
+ },
425
+ {
426
+ "epoch": 0.33,
427
+ "learning_rate": 4.967933226753913e-05,
428
+ "loss": 3.0407,
429
+ "step": 35000
430
+ },
431
+ {
432
+ "epoch": 0.33,
433
+ "learning_rate": 4.967468490909768e-05,
434
+ "loss": 3.0363,
435
+ "step": 35500
436
+ },
437
+ {
438
+ "epoch": 0.33,
439
+ "learning_rate": 4.967003755065621e-05,
440
+ "loss": 3.0204,
441
+ "step": 36000
442
+ },
443
+ {
444
+ "epoch": 0.34,
445
+ "learning_rate": 4.966539019221475e-05,
446
+ "loss": 3.0192,
447
+ "step": 36500
448
+ },
449
+ {
450
+ "epoch": 0.34,
451
+ "learning_rate": 4.966074283377329e-05,
452
+ "loss": 3.0018,
453
+ "step": 37000
454
+ },
455
+ {
456
+ "epoch": 0.35,
457
+ "learning_rate": 4.9656095475331824e-05,
458
+ "loss": 2.992,
459
+ "step": 37500
460
+ },
461
+ {
462
+ "epoch": 0.35,
463
+ "learning_rate": 4.9651448116890365e-05,
464
+ "loss": 2.9887,
465
+ "step": 38000
466
+ },
467
+ {
468
+ "epoch": 0.36,
469
+ "learning_rate": 4.96468007584489e-05,
470
+ "loss": 2.9729,
471
+ "step": 38500
472
+ },
473
+ {
474
+ "epoch": 0.36,
475
+ "learning_rate": 4.964215340000744e-05,
476
+ "loss": 2.9743,
477
+ "step": 39000
478
+ },
479
+ {
480
+ "epoch": 0.37,
481
+ "learning_rate": 4.963750604156598e-05,
482
+ "loss": 2.9579,
483
+ "step": 39500
484
+ },
485
+ {
486
+ "epoch": 0.37,
487
+ "learning_rate": 4.963285868312452e-05,
488
+ "loss": 2.9507,
489
+ "step": 40000
490
+ },
491
+ {
492
+ "epoch": 0.38,
493
+ "learning_rate": 4.9628211324683054e-05,
494
+ "loss": 2.9553,
495
+ "step": 40500
496
+ },
497
+ {
498
+ "epoch": 0.38,
499
+ "learning_rate": 4.9623563966241595e-05,
500
+ "loss": 2.9419,
501
+ "step": 41000
502
+ },
503
+ {
504
+ "epoch": 0.39,
505
+ "learning_rate": 4.961891660780013e-05,
506
+ "loss": 2.9302,
507
+ "step": 41500
508
+ },
509
+ {
510
+ "epoch": 0.39,
511
+ "learning_rate": 4.961426924935867e-05,
512
+ "loss": 2.934,
513
+ "step": 42000
514
+ },
515
+ {
516
+ "epoch": 0.39,
517
+ "learning_rate": 4.960962189091721e-05,
518
+ "loss": 2.9192,
519
+ "step": 42500
520
+ },
521
+ {
522
+ "epoch": 0.4,
523
+ "learning_rate": 4.960497453247575e-05,
524
+ "loss": 2.9089,
525
+ "step": 43000
526
+ },
527
+ {
528
+ "epoch": 0.4,
529
+ "learning_rate": 4.9600327174034283e-05,
530
+ "loss": 2.9131,
531
+ "step": 43500
532
+ },
533
+ {
534
+ "epoch": 0.41,
535
+ "learning_rate": 4.959567981559282e-05,
536
+ "loss": 2.8963,
537
+ "step": 44000
538
+ },
539
+ {
540
+ "epoch": 0.41,
541
+ "learning_rate": 4.959103245715136e-05,
542
+ "loss": 2.8968,
543
+ "step": 44500
544
+ },
545
+ {
546
+ "epoch": 0.42,
547
+ "learning_rate": 4.9586385098709895e-05,
548
+ "loss": 2.8909,
549
+ "step": 45000
550
+ },
551
+ {
552
+ "epoch": 0.42,
553
+ "learning_rate": 4.958173774026844e-05,
554
+ "loss": 2.8808,
555
+ "step": 45500
556
+ },
557
+ {
558
+ "epoch": 0.43,
559
+ "learning_rate": 4.957709038182697e-05,
560
+ "loss": 2.879,
561
+ "step": 46000
562
+ },
563
+ {
564
+ "epoch": 0.43,
565
+ "learning_rate": 4.957244302338551e-05,
566
+ "loss": 2.8674,
567
+ "step": 46500
568
+ },
569
+ {
570
+ "epoch": 0.44,
571
+ "learning_rate": 4.956779566494405e-05,
572
+ "loss": 2.8615,
573
+ "step": 47000
574
+ },
575
+ {
576
+ "epoch": 0.44,
577
+ "learning_rate": 4.956314830650259e-05,
578
+ "loss": 2.8619,
579
+ "step": 47500
580
+ },
581
+ {
582
+ "epoch": 0.45,
583
+ "learning_rate": 4.9558500948061125e-05,
584
+ "loss": 2.8642,
585
+ "step": 48000
586
+ },
587
+ {
588
+ "epoch": 0.45,
589
+ "learning_rate": 4.9553853589619667e-05,
590
+ "loss": 2.8486,
591
+ "step": 48500
592
+ },
593
+ {
594
+ "epoch": 0.46,
595
+ "learning_rate": 4.95492062311782e-05,
596
+ "loss": 2.8469,
597
+ "step": 49000
598
+ },
599
+ {
600
+ "epoch": 0.46,
601
+ "learning_rate": 4.954455887273674e-05,
602
+ "loss": 2.8438,
603
+ "step": 49500
604
+ },
605
+ {
606
+ "epoch": 0.46,
607
+ "learning_rate": 4.953991151429528e-05,
608
+ "loss": 2.832,
609
+ "step": 50000
610
+ },
611
+ {
612
+ "epoch": 0.47,
613
+ "learning_rate": 4.953526415585382e-05,
614
+ "loss": 2.8319,
615
+ "step": 50500
616
+ },
617
+ {
618
+ "epoch": 0.47,
619
+ "learning_rate": 4.9530616797412355e-05,
620
+ "loss": 2.8332,
621
+ "step": 51000
622
+ },
623
+ {
624
+ "epoch": 0.48,
625
+ "learning_rate": 4.952596943897089e-05,
626
+ "loss": 2.8333,
627
+ "step": 51500
628
+ },
629
+ {
630
+ "epoch": 0.48,
631
+ "learning_rate": 4.952132208052943e-05,
632
+ "loss": 2.8224,
633
+ "step": 52000
634
+ },
635
+ {
636
+ "epoch": 0.49,
637
+ "learning_rate": 4.9516674722087966e-05,
638
+ "loss": 2.8138,
639
+ "step": 52500
640
+ },
641
+ {
642
+ "epoch": 0.49,
643
+ "learning_rate": 4.951202736364651e-05,
644
+ "loss": 2.8129,
645
+ "step": 53000
646
+ },
647
+ {
648
+ "epoch": 0.5,
649
+ "learning_rate": 4.950738000520504e-05,
650
+ "loss": 2.8076,
651
+ "step": 53500
652
+ },
653
+ {
654
+ "epoch": 0.5,
655
+ "learning_rate": 4.9502732646763585e-05,
656
+ "loss": 2.7981,
657
+ "step": 54000
658
+ },
659
+ {
660
+ "epoch": 0.51,
661
+ "learning_rate": 4.949808528832212e-05,
662
+ "loss": 2.8029,
663
+ "step": 54500
664
+ },
665
+ {
666
+ "epoch": 0.51,
667
+ "learning_rate": 4.949343792988066e-05,
668
+ "loss": 2.7994,
669
+ "step": 55000
670
+ },
671
+ {
672
+ "epoch": 0.52,
673
+ "learning_rate": 4.9488790571439196e-05,
674
+ "loss": 2.793,
675
+ "step": 55500
676
+ },
677
+ {
678
+ "epoch": 0.52,
679
+ "learning_rate": 4.948414321299774e-05,
680
+ "loss": 2.7945,
681
+ "step": 56000
682
+ },
683
+ {
684
+ "epoch": 0.53,
685
+ "learning_rate": 4.947949585455627e-05,
686
+ "loss": 2.7819,
687
+ "step": 56500
688
+ },
689
+ {
690
+ "epoch": 0.53,
691
+ "learning_rate": 4.9474848496114814e-05,
692
+ "loss": 2.7854,
693
+ "step": 57000
694
+ },
695
+ {
696
+ "epoch": 0.53,
697
+ "learning_rate": 4.947020113767335e-05,
698
+ "loss": 2.7698,
699
+ "step": 57500
700
+ },
701
+ {
702
+ "epoch": 0.54,
703
+ "learning_rate": 4.946555377923189e-05,
704
+ "loss": 2.7791,
705
+ "step": 58000
706
+ },
707
+ {
708
+ "epoch": 0.54,
709
+ "learning_rate": 4.9460906420790426e-05,
710
+ "loss": 2.7766,
711
+ "step": 58500
712
+ },
713
+ {
714
+ "epoch": 0.55,
715
+ "learning_rate": 4.945625906234896e-05,
716
+ "loss": 2.7627,
717
+ "step": 59000
718
+ },
719
+ {
720
+ "epoch": 0.55,
721
+ "learning_rate": 4.94516117039075e-05,
722
+ "loss": 2.7705,
723
+ "step": 59500
724
+ },
725
+ {
726
+ "epoch": 0.56,
727
+ "learning_rate": 4.944696434546604e-05,
728
+ "loss": 2.7582,
729
+ "step": 60000
730
+ },
731
+ {
732
+ "epoch": 0.56,
733
+ "learning_rate": 4.944231698702458e-05,
734
+ "loss": 2.7632,
735
+ "step": 60500
736
+ },
737
+ {
738
+ "epoch": 0.57,
739
+ "learning_rate": 4.9437669628583114e-05,
740
+ "loss": 2.764,
741
+ "step": 61000
742
+ },
743
+ {
744
+ "epoch": 0.57,
745
+ "learning_rate": 4.9433022270141656e-05,
746
+ "loss": 2.753,
747
+ "step": 61500
748
+ },
749
+ {
750
+ "epoch": 0.58,
751
+ "learning_rate": 4.942837491170019e-05,
752
+ "loss": 2.7469,
753
+ "step": 62000
754
+ },
755
+ {
756
+ "epoch": 0.58,
757
+ "learning_rate": 4.942372755325873e-05,
758
+ "loss": 2.7506,
759
+ "step": 62500
760
+ },
761
+ {
762
+ "epoch": 0.59,
763
+ "learning_rate": 4.941908019481727e-05,
764
+ "loss": 2.746,
765
+ "step": 63000
766
+ },
767
+ {
768
+ "epoch": 0.59,
769
+ "learning_rate": 4.941443283637581e-05,
770
+ "loss": 2.7373,
771
+ "step": 63500
772
+ },
773
+ {
774
+ "epoch": 0.59,
775
+ "learning_rate": 4.9409785477934344e-05,
776
+ "loss": 2.7325,
777
+ "step": 64000
778
+ },
779
+ {
780
+ "epoch": 0.6,
781
+ "learning_rate": 4.9405138119492886e-05,
782
+ "loss": 2.7329,
783
+ "step": 64500
784
+ },
785
+ {
786
+ "epoch": 0.6,
787
+ "learning_rate": 4.940049076105142e-05,
788
+ "loss": 2.7313,
789
+ "step": 65000
790
+ },
791
+ {
792
+ "epoch": 0.61,
793
+ "learning_rate": 4.939584340260996e-05,
794
+ "loss": 2.728,
795
+ "step": 65500
796
+ },
797
+ {
798
+ "epoch": 0.61,
799
+ "learning_rate": 4.93911960441685e-05,
800
+ "loss": 2.7244,
801
+ "step": 66000
802
+ },
803
+ {
804
+ "epoch": 0.62,
805
+ "learning_rate": 4.938654868572703e-05,
806
+ "loss": 2.722,
807
+ "step": 66500
808
+ },
809
+ {
810
+ "epoch": 0.62,
811
+ "learning_rate": 4.9381901327285574e-05,
812
+ "loss": 2.7278,
813
+ "step": 67000
814
+ },
815
+ {
816
+ "epoch": 0.63,
817
+ "learning_rate": 4.937725396884411e-05,
818
+ "loss": 2.7148,
819
+ "step": 67500
820
+ },
821
+ {
822
+ "epoch": 0.63,
823
+ "learning_rate": 4.937260661040265e-05,
824
+ "loss": 2.7251,
825
+ "step": 68000
826
+ },
827
+ {
828
+ "epoch": 0.64,
829
+ "learning_rate": 4.9367959251961185e-05,
830
+ "loss": 2.7076,
831
+ "step": 68500
832
+ },
833
+ {
834
+ "epoch": 0.64,
835
+ "learning_rate": 4.936331189351973e-05,
836
+ "loss": 2.717,
837
+ "step": 69000
838
+ },
839
+ {
840
+ "epoch": 0.65,
841
+ "learning_rate": 4.935866453507826e-05,
842
+ "loss": 2.7075,
843
+ "step": 69500
844
+ },
845
+ {
846
+ "epoch": 0.65,
847
+ "learning_rate": 4.9354017176636804e-05,
848
+ "loss": 2.7132,
849
+ "step": 70000
850
+ },
851
+ {
852
+ "epoch": 0.66,
853
+ "learning_rate": 4.934936981819534e-05,
854
+ "loss": 2.7055,
855
+ "step": 70500
856
+ },
857
+ {
858
+ "epoch": 0.66,
859
+ "learning_rate": 4.934472245975388e-05,
860
+ "loss": 2.7106,
861
+ "step": 71000
862
+ },
863
+ {
864
+ "epoch": 0.66,
865
+ "learning_rate": 4.9340075101312415e-05,
866
+ "loss": 2.71,
867
+ "step": 71500
868
+ },
869
+ {
870
+ "epoch": 0.67,
871
+ "learning_rate": 4.933542774287096e-05,
872
+ "loss": 2.6932,
873
+ "step": 72000
874
+ },
875
+ {
876
+ "epoch": 0.67,
877
+ "learning_rate": 4.933078038442949e-05,
878
+ "loss": 2.7055,
879
+ "step": 72500
880
+ },
881
+ {
882
+ "epoch": 0.68,
883
+ "learning_rate": 4.9326133025988034e-05,
884
+ "loss": 2.7001,
885
+ "step": 73000
886
+ },
887
+ {
888
+ "epoch": 0.68,
889
+ "learning_rate": 4.932148566754657e-05,
890
+ "loss": 2.6966,
891
+ "step": 73500
892
+ },
893
+ {
894
+ "epoch": 0.69,
895
+ "learning_rate": 4.93168383091051e-05,
896
+ "loss": 2.6817,
897
+ "step": 74000
898
+ },
899
+ {
900
+ "epoch": 0.69,
901
+ "learning_rate": 4.9312190950663645e-05,
902
+ "loss": 2.6838,
903
+ "step": 74500
904
+ },
905
+ {
906
+ "epoch": 0.7,
907
+ "learning_rate": 4.930754359222218e-05,
908
+ "loss": 2.6867,
909
+ "step": 75000
910
+ },
911
+ {
912
+ "epoch": 0.7,
913
+ "learning_rate": 4.930289623378072e-05,
914
+ "loss": 2.6786,
915
+ "step": 75500
916
+ },
917
+ {
918
+ "epoch": 0.71,
919
+ "learning_rate": 4.929824887533926e-05,
920
+ "loss": 2.6769,
921
+ "step": 76000
922
+ },
923
+ {
924
+ "epoch": 0.71,
925
+ "learning_rate": 4.92936015168978e-05,
926
+ "loss": 2.668,
927
+ "step": 76500
928
+ },
929
+ {
930
+ "epoch": 0.72,
931
+ "learning_rate": 4.928895415845633e-05,
932
+ "loss": 2.6775,
933
+ "step": 77000
934
+ },
935
+ {
936
+ "epoch": 0.72,
937
+ "learning_rate": 4.9284306800014875e-05,
938
+ "loss": 2.6715,
939
+ "step": 77500
940
+ },
941
+ {
942
+ "epoch": 0.72,
943
+ "learning_rate": 4.927965944157341e-05,
944
+ "loss": 2.67,
945
+ "step": 78000
946
+ },
947
+ {
948
+ "epoch": 0.73,
949
+ "learning_rate": 4.927501208313195e-05,
950
+ "loss": 2.6713,
951
+ "step": 78500
952
+ },
953
+ {
954
+ "epoch": 0.73,
955
+ "learning_rate": 4.9270364724690486e-05,
956
+ "loss": 2.6632,
957
+ "step": 79000
958
+ },
959
+ {
960
+ "epoch": 0.74,
961
+ "learning_rate": 4.926571736624903e-05,
962
+ "loss": 2.6631,
963
+ "step": 79500
964
+ },
965
+ {
966
+ "epoch": 0.74,
967
+ "learning_rate": 4.926107000780756e-05,
968
+ "loss": 2.6666,
969
+ "step": 80000
970
+ },
971
+ {
972
+ "epoch": 0.75,
973
+ "learning_rate": 4.9256422649366105e-05,
974
+ "loss": 2.6637,
975
+ "step": 80500
976
+ },
977
+ {
978
+ "epoch": 0.75,
979
+ "learning_rate": 4.925177529092464e-05,
980
+ "loss": 2.6602,
981
+ "step": 81000
982
+ },
983
+ {
984
+ "epoch": 0.76,
985
+ "learning_rate": 4.924712793248318e-05,
986
+ "loss": 2.657,
987
+ "step": 81500
988
+ },
989
+ {
990
+ "epoch": 0.76,
991
+ "learning_rate": 4.9242480574041716e-05,
992
+ "loss": 2.6548,
993
+ "step": 82000
994
+ },
995
+ {
996
+ "epoch": 0.77,
997
+ "learning_rate": 4.923783321560025e-05,
998
+ "loss": 2.6514,
999
+ "step": 82500
1000
+ },
1001
+ {
1002
+ "epoch": 0.77,
1003
+ "learning_rate": 4.923318585715879e-05,
1004
+ "loss": 2.6485,
1005
+ "step": 83000
1006
+ },
1007
+ {
1008
+ "epoch": 0.78,
1009
+ "learning_rate": 4.922853849871733e-05,
1010
+ "loss": 2.6519,
1011
+ "step": 83500
1012
+ },
1013
+ {
1014
+ "epoch": 0.78,
1015
+ "learning_rate": 4.922389114027587e-05,
1016
+ "loss": 2.6462,
1017
+ "step": 84000
1018
+ },
1019
+ {
1020
+ "epoch": 0.79,
1021
+ "learning_rate": 4.9219243781834404e-05,
1022
+ "loss": 2.6415,
1023
+ "step": 84500
1024
+ },
1025
+ {
1026
+ "epoch": 0.79,
1027
+ "learning_rate": 4.9214596423392946e-05,
1028
+ "loss": 2.6441,
1029
+ "step": 85000
1030
+ },
1031
+ {
1032
+ "epoch": 0.79,
1033
+ "learning_rate": 4.920994906495148e-05,
1034
+ "loss": 2.6473,
1035
+ "step": 85500
1036
+ },
1037
+ {
1038
+ "epoch": 0.8,
1039
+ "learning_rate": 4.920530170651002e-05,
1040
+ "loss": 2.6343,
1041
+ "step": 86000
1042
+ },
1043
+ {
1044
+ "epoch": 0.8,
1045
+ "learning_rate": 4.920065434806856e-05,
1046
+ "loss": 2.634,
1047
+ "step": 86500
1048
+ },
1049
+ {
1050
+ "epoch": 0.81,
1051
+ "learning_rate": 4.91960069896271e-05,
1052
+ "loss": 2.6393,
1053
+ "step": 87000
1054
+ },
1055
+ {
1056
+ "epoch": 0.81,
1057
+ "learning_rate": 4.9191359631185634e-05,
1058
+ "loss": 2.6328,
1059
+ "step": 87500
1060
+ },
1061
+ {
1062
+ "epoch": 0.82,
1063
+ "learning_rate": 4.9186712272744176e-05,
1064
+ "loss": 2.6286,
1065
+ "step": 88000
1066
+ },
1067
+ {
1068
+ "epoch": 0.82,
1069
+ "learning_rate": 4.918206491430271e-05,
1070
+ "loss": 2.6346,
1071
+ "step": 88500
1072
+ },
1073
+ {
1074
+ "epoch": 0.83,
1075
+ "learning_rate": 4.917741755586125e-05,
1076
+ "loss": 2.6302,
1077
+ "step": 89000
1078
+ },
1079
+ {
1080
+ "epoch": 0.83,
1081
+ "learning_rate": 4.917277019741979e-05,
1082
+ "loss": 2.6394,
1083
+ "step": 89500
1084
+ },
1085
+ {
1086
+ "epoch": 0.84,
1087
+ "learning_rate": 4.916812283897832e-05,
1088
+ "loss": 2.6319,
1089
+ "step": 90000
1090
+ },
1091
+ {
1092
+ "epoch": 0.84,
1093
+ "learning_rate": 4.9163475480536864e-05,
1094
+ "loss": 2.6225,
1095
+ "step": 90500
1096
+ },
1097
+ {
1098
+ "epoch": 0.85,
1099
+ "learning_rate": 4.91588281220954e-05,
1100
+ "loss": 2.6239,
1101
+ "step": 91000
1102
+ },
1103
+ {
1104
+ "epoch": 0.85,
1105
+ "learning_rate": 4.915418076365394e-05,
1106
+ "loss": 2.6232,
1107
+ "step": 91500
1108
+ },
1109
+ {
1110
+ "epoch": 0.86,
1111
+ "learning_rate": 4.9149533405212476e-05,
1112
+ "loss": 2.6222,
1113
+ "step": 92000
1114
+ },
1115
+ {
1116
+ "epoch": 0.86,
1117
+ "learning_rate": 4.914488604677102e-05,
1118
+ "loss": 2.6206,
1119
+ "step": 92500
1120
+ },
1121
+ {
1122
+ "epoch": 0.86,
1123
+ "learning_rate": 4.914023868832955e-05,
1124
+ "loss": 2.6182,
1125
+ "step": 93000
1126
+ },
1127
+ {
1128
+ "epoch": 0.87,
1129
+ "learning_rate": 4.9135591329888094e-05,
1130
+ "loss": 2.6104,
1131
+ "step": 93500
1132
+ },
1133
+ {
1134
+ "epoch": 0.87,
1135
+ "learning_rate": 4.913094397144663e-05,
1136
+ "loss": 2.6129,
1137
+ "step": 94000
1138
+ },
1139
+ {
1140
+ "epoch": 0.88,
1141
+ "learning_rate": 4.912629661300517e-05,
1142
+ "loss": 2.606,
1143
+ "step": 94500
1144
+ },
1145
+ {
1146
+ "epoch": 0.88,
1147
+ "learning_rate": 4.9121649254563706e-05,
1148
+ "loss": 2.6154,
1149
+ "step": 95000
1150
+ },
1151
+ {
1152
+ "epoch": 0.89,
1153
+ "learning_rate": 4.911700189612225e-05,
1154
+ "loss": 2.6098,
1155
+ "step": 95500
1156
+ },
1157
+ {
1158
+ "epoch": 0.89,
1159
+ "learning_rate": 4.911235453768078e-05,
1160
+ "loss": 2.6139,
1161
+ "step": 96000
1162
+ },
1163
+ {
1164
+ "epoch": 0.9,
1165
+ "learning_rate": 4.9107707179239324e-05,
1166
+ "loss": 2.6098,
1167
+ "step": 96500
1168
+ },
1169
+ {
1170
+ "epoch": 0.9,
1171
+ "learning_rate": 4.910305982079786e-05,
1172
+ "loss": 2.601,
1173
+ "step": 97000
1174
+ },
1175
+ {
1176
+ "epoch": 0.91,
1177
+ "learning_rate": 4.9098412462356394e-05,
1178
+ "loss": 2.6003,
1179
+ "step": 97500
1180
+ },
1181
+ {
1182
+ "epoch": 0.91,
1183
+ "learning_rate": 4.9093765103914935e-05,
1184
+ "loss": 2.6065,
1185
+ "step": 98000
1186
+ },
1187
+ {
1188
+ "epoch": 0.92,
1189
+ "learning_rate": 4.908911774547347e-05,
1190
+ "loss": 2.6015,
1191
+ "step": 98500
1192
+ },
1193
+ {
1194
+ "epoch": 0.92,
1195
+ "learning_rate": 4.908447038703201e-05,
1196
+ "loss": 2.5993,
1197
+ "step": 99000
1198
+ },
1199
+ {
1200
+ "epoch": 0.92,
1201
+ "learning_rate": 4.907982302859055e-05,
1202
+ "loss": 2.6024,
1203
+ "step": 99500
1204
+ },
1205
+ {
1206
+ "epoch": 0.93,
1207
+ "learning_rate": 4.907517567014909e-05,
1208
+ "loss": 2.5903,
1209
+ "step": 100000
1210
+ },
1211
+ {
1212
+ "epoch": 0.93,
1213
+ "learning_rate": 4.9070528311707624e-05,
1214
+ "loss": 2.5874,
1215
+ "step": 100500
1216
+ },
1217
+ {
1218
+ "epoch": 0.94,
1219
+ "learning_rate": 4.9065880953266165e-05,
1220
+ "loss": 2.5897,
1221
+ "step": 101000
1222
+ },
1223
+ {
1224
+ "epoch": 0.94,
1225
+ "learning_rate": 4.90612335948247e-05,
1226
+ "loss": 2.5892,
1227
+ "step": 101500
1228
+ },
1229
+ {
1230
+ "epoch": 0.95,
1231
+ "learning_rate": 4.905658623638324e-05,
1232
+ "loss": 2.5923,
1233
+ "step": 102000
1234
+ },
1235
+ {
1236
+ "epoch": 0.95,
1237
+ "learning_rate": 4.905193887794178e-05,
1238
+ "loss": 2.5904,
1239
+ "step": 102500
1240
+ },
1241
+ {
1242
+ "epoch": 0.96,
1243
+ "learning_rate": 4.904729151950032e-05,
1244
+ "loss": 2.5814,
1245
+ "step": 103000
1246
+ },
1247
+ {
1248
+ "epoch": 0.96,
1249
+ "learning_rate": 4.9042644161058853e-05,
1250
+ "loss": 2.5814,
1251
+ "step": 103500
1252
+ },
1253
+ {
1254
+ "epoch": 0.97,
1255
+ "learning_rate": 4.9037996802617395e-05,
1256
+ "loss": 2.5822,
1257
+ "step": 104000
1258
+ },
1259
+ {
1260
+ "epoch": 0.97,
1261
+ "learning_rate": 4.903334944417593e-05,
1262
+ "loss": 2.5875,
1263
+ "step": 104500
1264
+ },
1265
+ {
1266
+ "epoch": 0.98,
1267
+ "learning_rate": 4.9028702085734465e-05,
1268
+ "loss": 2.5868,
1269
+ "step": 105000
1270
+ },
1271
+ {
1272
+ "epoch": 0.98,
1273
+ "learning_rate": 4.9024054727293013e-05,
1274
+ "loss": 2.5837,
1275
+ "step": 105500
1276
+ },
1277
+ {
1278
+ "epoch": 0.99,
1279
+ "learning_rate": 4.901940736885155e-05,
1280
+ "loss": 2.5794,
1281
+ "step": 106000
1282
+ },
1283
+ {
1284
+ "epoch": 0.99,
1285
+ "learning_rate": 4.901476001041009e-05,
1286
+ "loss": 2.5774,
1287
+ "step": 106500
1288
+ },
1289
+ {
1290
+ "epoch": 0.99,
1291
+ "learning_rate": 4.9010112651968625e-05,
1292
+ "loss": 2.5793,
1293
+ "step": 107000
1294
+ },
1295
+ {
1296
+ "epoch": 1.0,
1297
+ "learning_rate": 4.900546529352717e-05,
1298
+ "loss": 2.5743,
1299
+ "step": 107500
1300
+ },
1301
+ {
1302
+ "epoch": 1.0,
1303
+ "learning_rate": 4.90008179350857e-05,
1304
+ "loss": 2.5456,
1305
+ "step": 108000
1306
+ },
1307
+ {
1308
+ "epoch": 1.01,
1309
+ "learning_rate": 4.899617057664424e-05,
1310
+ "loss": 2.5337,
1311
+ "step": 108500
1312
+ },
1313
+ {
1314
+ "epoch": 1.01,
1315
+ "learning_rate": 4.899152321820278e-05,
1316
+ "loss": 2.542,
1317
+ "step": 109000
1318
+ },
1319
+ {
1320
+ "epoch": 1.02,
1321
+ "learning_rate": 4.898687585976131e-05,
1322
+ "loss": 2.5439,
1323
+ "step": 109500
1324
+ },
1325
+ {
1326
+ "epoch": 1.02,
1327
+ "learning_rate": 4.8982228501319855e-05,
1328
+ "loss": 2.5341,
1329
+ "step": 110000
1330
+ },
1331
+ {
1332
+ "epoch": 1.03,
1333
+ "learning_rate": 4.897758114287839e-05,
1334
+ "loss": 2.5415,
1335
+ "step": 110500
1336
+ },
1337
+ {
1338
+ "epoch": 1.03,
1339
+ "learning_rate": 4.897293378443693e-05,
1340
+ "loss": 2.5367,
1341
+ "step": 111000
1342
+ },
1343
+ {
1344
+ "epoch": 1.04,
1345
+ "learning_rate": 4.8968286425995466e-05,
1346
+ "loss": 2.5401,
1347
+ "step": 111500
1348
+ },
1349
+ {
1350
+ "epoch": 1.04,
1351
+ "learning_rate": 4.896363906755401e-05,
1352
+ "loss": 2.5385,
1353
+ "step": 112000
1354
+ },
1355
+ {
1356
+ "epoch": 1.05,
1357
+ "learning_rate": 4.895899170911254e-05,
1358
+ "loss": 2.5412,
1359
+ "step": 112500
1360
+ },
1361
+ {
1362
+ "epoch": 1.05,
1363
+ "learning_rate": 4.8954344350671085e-05,
1364
+ "loss": 2.5451,
1365
+ "step": 113000
1366
+ },
1367
+ {
1368
+ "epoch": 1.05,
1369
+ "learning_rate": 4.894969699222962e-05,
1370
+ "loss": 2.5377,
1371
+ "step": 113500
1372
+ },
1373
+ {
1374
+ "epoch": 1.06,
1375
+ "learning_rate": 4.894504963378816e-05,
1376
+ "loss": 2.5349,
1377
+ "step": 114000
1378
+ },
1379
+ {
1380
+ "epoch": 1.06,
1381
+ "learning_rate": 4.8940402275346696e-05,
1382
+ "loss": 2.5364,
1383
+ "step": 114500
1384
+ },
1385
+ {
1386
+ "epoch": 1.07,
1387
+ "learning_rate": 4.893575491690524e-05,
1388
+ "loss": 2.5315,
1389
+ "step": 115000
1390
+ }
1391
+ ],
1392
+ "logging_steps": 500,
1393
+ "max_steps": 5379900,
1394
+ "num_input_tokens_seen": 0,
1395
+ "num_train_epochs": 50,
1396
+ "save_steps": 1000,
1397
+ "total_flos": 1.46391313022976e+19,
1398
+ "train_batch_size": 7,
1399
+ "trial_name": null,
1400
+ "trial_params": null
1401
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:621dab73aa3142c59d86181d90695227eeb17a81569693b1853db1b36f4692c0
3
+ size 4728
vocab.json ADDED
The diff for this file is too large to render. See raw diff