Muhammad Khalifa commited on
Commit
2824cc3
1 Parent(s): f633d8a

add llama-13b text

Browse files
text_models/llama-13b/best_model/adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "base_model_name_or_path": "huggyllama/llama-13b",
3
+ "bias": "none",
4
+ "enable_lora": null,
5
+ "fan_in_fan_out": false,
6
+ "inference_mode": true,
7
+ "init_lora_weights": true,
8
+ "lora_alpha": 16,
9
+ "lora_dropout": 0.05,
10
+ "merge_weights": false,
11
+ "modules_to_save": null,
12
+ "peft_type": "LORA",
13
+ "r": 16,
14
+ "target_modules": [
15
+ "q_proj",
16
+ "k_proj",
17
+ "v_proj",
18
+ "o_proj"
19
+ ],
20
+ "task_type": "CAUSAL_LM"
21
+ }
text_models/llama-13b/best_model/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ff531271dba892e7b7484234e0fccc927b4053b6db1aaaecd56dcaf62d2a6084
3
+ size 104973389
text_models/llama-13b/best_model/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:86f13d83336121166f375c5b0995f9c31fc31e2a55150bfdd45bf10a34f9e096
3
+ size 209984517
text_models/llama-13b/best_model/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0a288826c677385768968fe0aa5bf2f6a3d2ea16a9a025fee183feb0e39d9e9e
3
+ size 14575
text_models/llama-13b/best_model/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1c5ac04ba47cdd731ebd5689bdc424fd204bd0952c9ed071d8609381ffc0d569
3
+ size 627
text_models/llama-13b/best_model/trainer_state.json ADDED
@@ -0,0 +1,1939 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 1.0316544771194458,
3
+ "best_model_checkpoint": "checkpoints/instrucode/text_models/llama-13b/checkpoint-3000",
4
+ "epoch": 0.9599232061435086,
5
+ "eval_steps": 200,
6
+ "global_step": 3000,
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": 2.3999999999999997e-05,
14
+ "loss": 1.9718,
15
+ "step": 10
16
+ },
17
+ {
18
+ "epoch": 0.01,
19
+ "learning_rate": 5.399999999999999e-05,
20
+ "loss": 1.9714,
21
+ "step": 20
22
+ },
23
+ {
24
+ "epoch": 0.01,
25
+ "learning_rate": 8.4e-05,
26
+ "loss": 1.6976,
27
+ "step": 30
28
+ },
29
+ {
30
+ "epoch": 0.01,
31
+ "learning_rate": 0.00011399999999999999,
32
+ "loss": 1.4059,
33
+ "step": 40
34
+ },
35
+ {
36
+ "epoch": 0.02,
37
+ "learning_rate": 0.00014399999999999998,
38
+ "loss": 1.1741,
39
+ "step": 50
40
+ },
41
+ {
42
+ "epoch": 0.02,
43
+ "learning_rate": 0.00017399999999999997,
44
+ "loss": 1.1727,
45
+ "step": 60
46
+ },
47
+ {
48
+ "epoch": 0.02,
49
+ "learning_rate": 0.000204,
50
+ "loss": 1.1187,
51
+ "step": 70
52
+ },
53
+ {
54
+ "epoch": 0.03,
55
+ "learning_rate": 0.000234,
56
+ "loss": 1.1282,
57
+ "step": 80
58
+ },
59
+ {
60
+ "epoch": 0.03,
61
+ "learning_rate": 0.00026399999999999997,
62
+ "loss": 1.1329,
63
+ "step": 90
64
+ },
65
+ {
66
+ "epoch": 0.03,
67
+ "learning_rate": 0.000294,
68
+ "loss": 1.0721,
69
+ "step": 100
70
+ },
71
+ {
72
+ "epoch": 0.04,
73
+ "learning_rate": 0.0002998454106280193,
74
+ "loss": 1.118,
75
+ "step": 110
76
+ },
77
+ {
78
+ "epoch": 0.04,
79
+ "learning_rate": 0.00029965217391304344,
80
+ "loss": 1.1217,
81
+ "step": 120
82
+ },
83
+ {
84
+ "epoch": 0.04,
85
+ "learning_rate": 0.0002994589371980676,
86
+ "loss": 1.1033,
87
+ "step": 130
88
+ },
89
+ {
90
+ "epoch": 0.04,
91
+ "learning_rate": 0.0002992657004830917,
92
+ "loss": 1.1104,
93
+ "step": 140
94
+ },
95
+ {
96
+ "epoch": 0.05,
97
+ "learning_rate": 0.0002990724637681159,
98
+ "loss": 1.078,
99
+ "step": 150
100
+ },
101
+ {
102
+ "epoch": 0.05,
103
+ "learning_rate": 0.00029887922705314007,
104
+ "loss": 1.0807,
105
+ "step": 160
106
+ },
107
+ {
108
+ "epoch": 0.05,
109
+ "learning_rate": 0.0002986859903381642,
110
+ "loss": 1.0875,
111
+ "step": 170
112
+ },
113
+ {
114
+ "epoch": 0.06,
115
+ "learning_rate": 0.00029849275362318835,
116
+ "loss": 1.0463,
117
+ "step": 180
118
+ },
119
+ {
120
+ "epoch": 0.06,
121
+ "learning_rate": 0.00029829951690821255,
122
+ "loss": 1.1217,
123
+ "step": 190
124
+ },
125
+ {
126
+ "epoch": 0.06,
127
+ "learning_rate": 0.0002981062801932367,
128
+ "loss": 1.0452,
129
+ "step": 200
130
+ },
131
+ {
132
+ "epoch": 0.06,
133
+ "eval_loss": 1.0776479244232178,
134
+ "eval_runtime": 197.5388,
135
+ "eval_samples_per_second": 10.125,
136
+ "eval_steps_per_second": 1.266,
137
+ "step": 200
138
+ },
139
+ {
140
+ "epoch": 0.07,
141
+ "learning_rate": 0.00029791304347826083,
142
+ "loss": 1.0228,
143
+ "step": 210
144
+ },
145
+ {
146
+ "epoch": 0.07,
147
+ "learning_rate": 0.000297719806763285,
148
+ "loss": 1.092,
149
+ "step": 220
150
+ },
151
+ {
152
+ "epoch": 0.07,
153
+ "learning_rate": 0.0002975265700483091,
154
+ "loss": 1.104,
155
+ "step": 230
156
+ },
157
+ {
158
+ "epoch": 0.08,
159
+ "learning_rate": 0.0002973333333333333,
160
+ "loss": 1.0577,
161
+ "step": 240
162
+ },
163
+ {
164
+ "epoch": 0.08,
165
+ "learning_rate": 0.00029714009661835746,
166
+ "loss": 1.0892,
167
+ "step": 250
168
+ },
169
+ {
170
+ "epoch": 0.08,
171
+ "learning_rate": 0.0002969468599033816,
172
+ "loss": 1.0915,
173
+ "step": 260
174
+ },
175
+ {
176
+ "epoch": 0.09,
177
+ "learning_rate": 0.00029675362318840574,
178
+ "loss": 1.0616,
179
+ "step": 270
180
+ },
181
+ {
182
+ "epoch": 0.09,
183
+ "learning_rate": 0.0002965603864734299,
184
+ "loss": 1.071,
185
+ "step": 280
186
+ },
187
+ {
188
+ "epoch": 0.09,
189
+ "learning_rate": 0.0002963671497584541,
190
+ "loss": 1.08,
191
+ "step": 290
192
+ },
193
+ {
194
+ "epoch": 0.1,
195
+ "learning_rate": 0.0002961739130434782,
196
+ "loss": 1.0947,
197
+ "step": 300
198
+ },
199
+ {
200
+ "epoch": 0.1,
201
+ "learning_rate": 0.00029598067632850237,
202
+ "loss": 1.0547,
203
+ "step": 310
204
+ },
205
+ {
206
+ "epoch": 0.1,
207
+ "learning_rate": 0.0002957874396135265,
208
+ "loss": 1.0723,
209
+ "step": 320
210
+ },
211
+ {
212
+ "epoch": 0.11,
213
+ "learning_rate": 0.0002955942028985507,
214
+ "loss": 1.0367,
215
+ "step": 330
216
+ },
217
+ {
218
+ "epoch": 0.11,
219
+ "learning_rate": 0.00029540096618357485,
220
+ "loss": 1.0349,
221
+ "step": 340
222
+ },
223
+ {
224
+ "epoch": 0.11,
225
+ "learning_rate": 0.000295207729468599,
226
+ "loss": 1.0507,
227
+ "step": 350
228
+ },
229
+ {
230
+ "epoch": 0.12,
231
+ "learning_rate": 0.00029501449275362313,
232
+ "loss": 1.0568,
233
+ "step": 360
234
+ },
235
+ {
236
+ "epoch": 0.12,
237
+ "learning_rate": 0.0002948212560386473,
238
+ "loss": 1.0861,
239
+ "step": 370
240
+ },
241
+ {
242
+ "epoch": 0.12,
243
+ "learning_rate": 0.0002946280193236715,
244
+ "loss": 1.0287,
245
+ "step": 380
246
+ },
247
+ {
248
+ "epoch": 0.12,
249
+ "learning_rate": 0.0002944347826086956,
250
+ "loss": 1.1086,
251
+ "step": 390
252
+ },
253
+ {
254
+ "epoch": 0.13,
255
+ "learning_rate": 0.00029424154589371976,
256
+ "loss": 1.0916,
257
+ "step": 400
258
+ },
259
+ {
260
+ "epoch": 0.13,
261
+ "eval_loss": 1.0622169971466064,
262
+ "eval_runtime": 197.2041,
263
+ "eval_samples_per_second": 10.142,
264
+ "eval_steps_per_second": 1.268,
265
+ "step": 400
266
+ },
267
+ {
268
+ "epoch": 0.13,
269
+ "learning_rate": 0.0002940483091787439,
270
+ "loss": 1.075,
271
+ "step": 410
272
+ },
273
+ {
274
+ "epoch": 0.13,
275
+ "learning_rate": 0.0002938550724637681,
276
+ "loss": 1.0899,
277
+ "step": 420
278
+ },
279
+ {
280
+ "epoch": 0.14,
281
+ "learning_rate": 0.00029366183574879224,
282
+ "loss": 1.0565,
283
+ "step": 430
284
+ },
285
+ {
286
+ "epoch": 0.14,
287
+ "learning_rate": 0.0002934685990338164,
288
+ "loss": 1.0414,
289
+ "step": 440
290
+ },
291
+ {
292
+ "epoch": 0.14,
293
+ "learning_rate": 0.0002932753623188405,
294
+ "loss": 1.1033,
295
+ "step": 450
296
+ },
297
+ {
298
+ "epoch": 0.15,
299
+ "learning_rate": 0.00029308212560386467,
300
+ "loss": 1.0518,
301
+ "step": 460
302
+ },
303
+ {
304
+ "epoch": 0.15,
305
+ "learning_rate": 0.00029288888888888886,
306
+ "loss": 1.1324,
307
+ "step": 470
308
+ },
309
+ {
310
+ "epoch": 0.15,
311
+ "learning_rate": 0.000292695652173913,
312
+ "loss": 1.0275,
313
+ "step": 480
314
+ },
315
+ {
316
+ "epoch": 0.16,
317
+ "learning_rate": 0.00029250241545893715,
318
+ "loss": 1.029,
319
+ "step": 490
320
+ },
321
+ {
322
+ "epoch": 0.16,
323
+ "learning_rate": 0.0002923091787439613,
324
+ "loss": 1.0648,
325
+ "step": 500
326
+ },
327
+ {
328
+ "epoch": 0.16,
329
+ "learning_rate": 0.0002921159420289855,
330
+ "loss": 1.0746,
331
+ "step": 510
332
+ },
333
+ {
334
+ "epoch": 0.17,
335
+ "learning_rate": 0.00029192270531400963,
336
+ "loss": 1.0845,
337
+ "step": 520
338
+ },
339
+ {
340
+ "epoch": 0.17,
341
+ "learning_rate": 0.0002917294685990338,
342
+ "loss": 1.0186,
343
+ "step": 530
344
+ },
345
+ {
346
+ "epoch": 0.17,
347
+ "learning_rate": 0.0002915362318840579,
348
+ "loss": 1.0984,
349
+ "step": 540
350
+ },
351
+ {
352
+ "epoch": 0.18,
353
+ "learning_rate": 0.0002913429951690821,
354
+ "loss": 1.0706,
355
+ "step": 550
356
+ },
357
+ {
358
+ "epoch": 0.18,
359
+ "learning_rate": 0.00029114975845410626,
360
+ "loss": 1.0184,
361
+ "step": 560
362
+ },
363
+ {
364
+ "epoch": 0.18,
365
+ "learning_rate": 0.0002909565217391304,
366
+ "loss": 1.0686,
367
+ "step": 570
368
+ },
369
+ {
370
+ "epoch": 0.19,
371
+ "learning_rate": 0.00029076328502415454,
372
+ "loss": 1.0391,
373
+ "step": 580
374
+ },
375
+ {
376
+ "epoch": 0.19,
377
+ "learning_rate": 0.0002905700483091787,
378
+ "loss": 1.0616,
379
+ "step": 590
380
+ },
381
+ {
382
+ "epoch": 0.19,
383
+ "learning_rate": 0.0002903768115942029,
384
+ "loss": 1.0565,
385
+ "step": 600
386
+ },
387
+ {
388
+ "epoch": 0.19,
389
+ "eval_loss": 1.0566095113754272,
390
+ "eval_runtime": 196.6646,
391
+ "eval_samples_per_second": 10.17,
392
+ "eval_steps_per_second": 1.271,
393
+ "step": 600
394
+ },
395
+ {
396
+ "epoch": 0.2,
397
+ "learning_rate": 0.000290183574879227,
398
+ "loss": 1.0609,
399
+ "step": 610
400
+ },
401
+ {
402
+ "epoch": 0.2,
403
+ "learning_rate": 0.00028999033816425117,
404
+ "loss": 1.0633,
405
+ "step": 620
406
+ },
407
+ {
408
+ "epoch": 0.2,
409
+ "learning_rate": 0.0002897971014492753,
410
+ "loss": 1.0049,
411
+ "step": 630
412
+ },
413
+ {
414
+ "epoch": 0.2,
415
+ "learning_rate": 0.0002896038647342995,
416
+ "loss": 0.9926,
417
+ "step": 640
418
+ },
419
+ {
420
+ "epoch": 0.21,
421
+ "learning_rate": 0.00028941062801932365,
422
+ "loss": 1.0949,
423
+ "step": 650
424
+ },
425
+ {
426
+ "epoch": 0.21,
427
+ "learning_rate": 0.0002892173913043478,
428
+ "loss": 1.0686,
429
+ "step": 660
430
+ },
431
+ {
432
+ "epoch": 0.21,
433
+ "learning_rate": 0.00028902415458937193,
434
+ "loss": 1.0152,
435
+ "step": 670
436
+ },
437
+ {
438
+ "epoch": 0.22,
439
+ "learning_rate": 0.0002888309178743961,
440
+ "loss": 1.0462,
441
+ "step": 680
442
+ },
443
+ {
444
+ "epoch": 0.22,
445
+ "learning_rate": 0.00028863768115942027,
446
+ "loss": 1.0438,
447
+ "step": 690
448
+ },
449
+ {
450
+ "epoch": 0.22,
451
+ "learning_rate": 0.0002884444444444444,
452
+ "loss": 1.0124,
453
+ "step": 700
454
+ },
455
+ {
456
+ "epoch": 0.23,
457
+ "learning_rate": 0.00028825120772946856,
458
+ "loss": 0.9959,
459
+ "step": 710
460
+ },
461
+ {
462
+ "epoch": 0.23,
463
+ "learning_rate": 0.0002880579710144927,
464
+ "loss": 1.0467,
465
+ "step": 720
466
+ },
467
+ {
468
+ "epoch": 0.23,
469
+ "learning_rate": 0.0002878647342995169,
470
+ "loss": 1.0502,
471
+ "step": 730
472
+ },
473
+ {
474
+ "epoch": 0.24,
475
+ "learning_rate": 0.00028767149758454104,
476
+ "loss": 1.0763,
477
+ "step": 740
478
+ },
479
+ {
480
+ "epoch": 0.24,
481
+ "learning_rate": 0.0002874782608695652,
482
+ "loss": 1.0821,
483
+ "step": 750
484
+ },
485
+ {
486
+ "epoch": 0.24,
487
+ "learning_rate": 0.0002872850241545893,
488
+ "loss": 1.0667,
489
+ "step": 760
490
+ },
491
+ {
492
+ "epoch": 0.25,
493
+ "learning_rate": 0.00028709178743961347,
494
+ "loss": 1.0604,
495
+ "step": 770
496
+ },
497
+ {
498
+ "epoch": 0.25,
499
+ "learning_rate": 0.00028689855072463766,
500
+ "loss": 1.0555,
501
+ "step": 780
502
+ },
503
+ {
504
+ "epoch": 0.25,
505
+ "learning_rate": 0.0002867053140096618,
506
+ "loss": 1.0861,
507
+ "step": 790
508
+ },
509
+ {
510
+ "epoch": 0.26,
511
+ "learning_rate": 0.00028651207729468595,
512
+ "loss": 1.0458,
513
+ "step": 800
514
+ },
515
+ {
516
+ "epoch": 0.26,
517
+ "eval_loss": 1.0510022640228271,
518
+ "eval_runtime": 196.1318,
519
+ "eval_samples_per_second": 10.197,
520
+ "eval_steps_per_second": 1.275,
521
+ "step": 800
522
+ },
523
+ {
524
+ "epoch": 0.26,
525
+ "learning_rate": 0.0002863188405797101,
526
+ "loss": 1.0542,
527
+ "step": 810
528
+ },
529
+ {
530
+ "epoch": 0.26,
531
+ "learning_rate": 0.0002861256038647343,
532
+ "loss": 1.0783,
533
+ "step": 820
534
+ },
535
+ {
536
+ "epoch": 0.27,
537
+ "learning_rate": 0.00028593236714975843,
538
+ "loss": 1.0668,
539
+ "step": 830
540
+ },
541
+ {
542
+ "epoch": 0.27,
543
+ "learning_rate": 0.00028573913043478257,
544
+ "loss": 1.0369,
545
+ "step": 840
546
+ },
547
+ {
548
+ "epoch": 0.27,
549
+ "learning_rate": 0.0002855458937198067,
550
+ "loss": 1.0321,
551
+ "step": 850
552
+ },
553
+ {
554
+ "epoch": 0.28,
555
+ "learning_rate": 0.0002853526570048309,
556
+ "loss": 1.0515,
557
+ "step": 860
558
+ },
559
+ {
560
+ "epoch": 0.28,
561
+ "learning_rate": 0.00028515942028985505,
562
+ "loss": 1.0129,
563
+ "step": 870
564
+ },
565
+ {
566
+ "epoch": 0.28,
567
+ "learning_rate": 0.0002849661835748792,
568
+ "loss": 1.0042,
569
+ "step": 880
570
+ },
571
+ {
572
+ "epoch": 0.28,
573
+ "learning_rate": 0.00028477294685990334,
574
+ "loss": 1.0086,
575
+ "step": 890
576
+ },
577
+ {
578
+ "epoch": 0.29,
579
+ "learning_rate": 0.0002845797101449275,
580
+ "loss": 1.0555,
581
+ "step": 900
582
+ },
583
+ {
584
+ "epoch": 0.29,
585
+ "learning_rate": 0.0002843864734299517,
586
+ "loss": 0.9976,
587
+ "step": 910
588
+ },
589
+ {
590
+ "epoch": 0.29,
591
+ "learning_rate": 0.0002841932367149758,
592
+ "loss": 1.0679,
593
+ "step": 920
594
+ },
595
+ {
596
+ "epoch": 0.3,
597
+ "learning_rate": 0.00028399999999999996,
598
+ "loss": 1.0039,
599
+ "step": 930
600
+ },
601
+ {
602
+ "epoch": 0.3,
603
+ "learning_rate": 0.0002838067632850241,
604
+ "loss": 1.0254,
605
+ "step": 940
606
+ },
607
+ {
608
+ "epoch": 0.3,
609
+ "learning_rate": 0.0002836135265700483,
610
+ "loss": 1.0109,
611
+ "step": 950
612
+ },
613
+ {
614
+ "epoch": 0.31,
615
+ "learning_rate": 0.00028342028985507245,
616
+ "loss": 1.0158,
617
+ "step": 960
618
+ },
619
+ {
620
+ "epoch": 0.31,
621
+ "learning_rate": 0.0002832270531400966,
622
+ "loss": 1.0165,
623
+ "step": 970
624
+ },
625
+ {
626
+ "epoch": 0.31,
627
+ "learning_rate": 0.00028303381642512073,
628
+ "loss": 1.031,
629
+ "step": 980
630
+ },
631
+ {
632
+ "epoch": 0.32,
633
+ "learning_rate": 0.0002828405797101449,
634
+ "loss": 1.0534,
635
+ "step": 990
636
+ },
637
+ {
638
+ "epoch": 0.32,
639
+ "learning_rate": 0.00028264734299516907,
640
+ "loss": 1.066,
641
+ "step": 1000
642
+ },
643
+ {
644
+ "epoch": 0.32,
645
+ "eval_loss": 1.047478199005127,
646
+ "eval_runtime": 196.1193,
647
+ "eval_samples_per_second": 10.198,
648
+ "eval_steps_per_second": 1.275,
649
+ "step": 1000
650
+ },
651
+ {
652
+ "epoch": 0.32,
653
+ "learning_rate": 0.0002824541062801932,
654
+ "loss": 1.0527,
655
+ "step": 1010
656
+ },
657
+ {
658
+ "epoch": 0.33,
659
+ "learning_rate": 0.00028226086956521736,
660
+ "loss": 1.0677,
661
+ "step": 1020
662
+ },
663
+ {
664
+ "epoch": 0.33,
665
+ "learning_rate": 0.0002820676328502415,
666
+ "loss": 1.0409,
667
+ "step": 1030
668
+ },
669
+ {
670
+ "epoch": 0.33,
671
+ "learning_rate": 0.0002818743961352657,
672
+ "loss": 1.0414,
673
+ "step": 1040
674
+ },
675
+ {
676
+ "epoch": 0.34,
677
+ "learning_rate": 0.00028168115942028984,
678
+ "loss": 1.0466,
679
+ "step": 1050
680
+ },
681
+ {
682
+ "epoch": 0.34,
683
+ "learning_rate": 0.000281487922705314,
684
+ "loss": 1.0357,
685
+ "step": 1060
686
+ },
687
+ {
688
+ "epoch": 0.34,
689
+ "learning_rate": 0.0002812946859903381,
690
+ "loss": 1.0418,
691
+ "step": 1070
692
+ },
693
+ {
694
+ "epoch": 0.35,
695
+ "learning_rate": 0.00028110144927536227,
696
+ "loss": 1.0796,
697
+ "step": 1080
698
+ },
699
+ {
700
+ "epoch": 0.35,
701
+ "learning_rate": 0.00028090821256038646,
702
+ "loss": 1.0579,
703
+ "step": 1090
704
+ },
705
+ {
706
+ "epoch": 0.35,
707
+ "learning_rate": 0.0002807149758454106,
708
+ "loss": 1.0631,
709
+ "step": 1100
710
+ },
711
+ {
712
+ "epoch": 0.36,
713
+ "learning_rate": 0.00028052173913043475,
714
+ "loss": 1.0221,
715
+ "step": 1110
716
+ },
717
+ {
718
+ "epoch": 0.36,
719
+ "learning_rate": 0.0002803285024154589,
720
+ "loss": 1.049,
721
+ "step": 1120
722
+ },
723
+ {
724
+ "epoch": 0.36,
725
+ "learning_rate": 0.0002801352657004831,
726
+ "loss": 1.0192,
727
+ "step": 1130
728
+ },
729
+ {
730
+ "epoch": 0.36,
731
+ "learning_rate": 0.00027994202898550723,
732
+ "loss": 0.9978,
733
+ "step": 1140
734
+ },
735
+ {
736
+ "epoch": 0.37,
737
+ "learning_rate": 0.00027974879227053137,
738
+ "loss": 1.0044,
739
+ "step": 1150
740
+ },
741
+ {
742
+ "epoch": 0.37,
743
+ "learning_rate": 0.0002795555555555555,
744
+ "loss": 1.0802,
745
+ "step": 1160
746
+ },
747
+ {
748
+ "epoch": 0.37,
749
+ "learning_rate": 0.0002793623188405797,
750
+ "loss": 1.1231,
751
+ "step": 1170
752
+ },
753
+ {
754
+ "epoch": 0.38,
755
+ "learning_rate": 0.00027916908212560385,
756
+ "loss": 1.0323,
757
+ "step": 1180
758
+ },
759
+ {
760
+ "epoch": 0.38,
761
+ "learning_rate": 0.000278975845410628,
762
+ "loss": 1.0908,
763
+ "step": 1190
764
+ },
765
+ {
766
+ "epoch": 0.38,
767
+ "learning_rate": 0.00027878260869565214,
768
+ "loss": 1.0456,
769
+ "step": 1200
770
+ },
771
+ {
772
+ "epoch": 0.38,
773
+ "eval_loss": 1.0448572635650635,
774
+ "eval_runtime": 197.1129,
775
+ "eval_samples_per_second": 10.146,
776
+ "eval_steps_per_second": 1.268,
777
+ "step": 1200
778
+ },
779
+ {
780
+ "epoch": 0.39,
781
+ "learning_rate": 0.0002785893719806763,
782
+ "loss": 1.0119,
783
+ "step": 1210
784
+ },
785
+ {
786
+ "epoch": 0.39,
787
+ "learning_rate": 0.0002783961352657005,
788
+ "loss": 1.0101,
789
+ "step": 1220
790
+ },
791
+ {
792
+ "epoch": 0.39,
793
+ "learning_rate": 0.0002782028985507246,
794
+ "loss": 1.0481,
795
+ "step": 1230
796
+ },
797
+ {
798
+ "epoch": 0.4,
799
+ "learning_rate": 0.00027800966183574876,
800
+ "loss": 1.0478,
801
+ "step": 1240
802
+ },
803
+ {
804
+ "epoch": 0.4,
805
+ "learning_rate": 0.0002778164251207729,
806
+ "loss": 1.0388,
807
+ "step": 1250
808
+ },
809
+ {
810
+ "epoch": 0.4,
811
+ "learning_rate": 0.0002776231884057971,
812
+ "loss": 1.0566,
813
+ "step": 1260
814
+ },
815
+ {
816
+ "epoch": 0.41,
817
+ "learning_rate": 0.00027742995169082124,
818
+ "loss": 1.084,
819
+ "step": 1270
820
+ },
821
+ {
822
+ "epoch": 0.41,
823
+ "learning_rate": 0.0002772367149758454,
824
+ "loss": 1.0515,
825
+ "step": 1280
826
+ },
827
+ {
828
+ "epoch": 0.41,
829
+ "learning_rate": 0.00027704347826086953,
830
+ "loss": 1.053,
831
+ "step": 1290
832
+ },
833
+ {
834
+ "epoch": 0.42,
835
+ "learning_rate": 0.00027685024154589367,
836
+ "loss": 1.0276,
837
+ "step": 1300
838
+ },
839
+ {
840
+ "epoch": 0.42,
841
+ "learning_rate": 0.00027665700483091787,
842
+ "loss": 1.0394,
843
+ "step": 1310
844
+ },
845
+ {
846
+ "epoch": 0.42,
847
+ "learning_rate": 0.000276463768115942,
848
+ "loss": 0.9991,
849
+ "step": 1320
850
+ },
851
+ {
852
+ "epoch": 0.43,
853
+ "learning_rate": 0.00027627053140096615,
854
+ "loss": 1.0272,
855
+ "step": 1330
856
+ },
857
+ {
858
+ "epoch": 0.43,
859
+ "learning_rate": 0.0002760772946859903,
860
+ "loss": 1.0684,
861
+ "step": 1340
862
+ },
863
+ {
864
+ "epoch": 0.43,
865
+ "learning_rate": 0.0002758840579710145,
866
+ "loss": 1.0865,
867
+ "step": 1350
868
+ },
869
+ {
870
+ "epoch": 0.44,
871
+ "learning_rate": 0.00027569082125603864,
872
+ "loss": 1.0375,
873
+ "step": 1360
874
+ },
875
+ {
876
+ "epoch": 0.44,
877
+ "learning_rate": 0.0002754975845410628,
878
+ "loss": 1.0922,
879
+ "step": 1370
880
+ },
881
+ {
882
+ "epoch": 0.44,
883
+ "learning_rate": 0.0002753043478260869,
884
+ "loss": 1.0611,
885
+ "step": 1380
886
+ },
887
+ {
888
+ "epoch": 0.44,
889
+ "learning_rate": 0.00027511111111111106,
890
+ "loss": 0.9878,
891
+ "step": 1390
892
+ },
893
+ {
894
+ "epoch": 0.45,
895
+ "learning_rate": 0.00027491787439613526,
896
+ "loss": 1.0658,
897
+ "step": 1400
898
+ },
899
+ {
900
+ "epoch": 0.45,
901
+ "eval_loss": 1.0436240434646606,
902
+ "eval_runtime": 197.0464,
903
+ "eval_samples_per_second": 10.15,
904
+ "eval_steps_per_second": 1.269,
905
+ "step": 1400
906
+ },
907
+ {
908
+ "epoch": 0.45,
909
+ "learning_rate": 0.0002747246376811594,
910
+ "loss": 1.0201,
911
+ "step": 1410
912
+ },
913
+ {
914
+ "epoch": 0.45,
915
+ "learning_rate": 0.00027453140096618355,
916
+ "loss": 1.0537,
917
+ "step": 1420
918
+ },
919
+ {
920
+ "epoch": 0.46,
921
+ "learning_rate": 0.0002743381642512077,
922
+ "loss": 1.051,
923
+ "step": 1430
924
+ },
925
+ {
926
+ "epoch": 0.46,
927
+ "learning_rate": 0.0002741449275362319,
928
+ "loss": 1.0028,
929
+ "step": 1440
930
+ },
931
+ {
932
+ "epoch": 0.46,
933
+ "learning_rate": 0.00027395169082125603,
934
+ "loss": 1.0521,
935
+ "step": 1450
936
+ },
937
+ {
938
+ "epoch": 0.47,
939
+ "learning_rate": 0.00027375845410628017,
940
+ "loss": 1.0732,
941
+ "step": 1460
942
+ },
943
+ {
944
+ "epoch": 0.47,
945
+ "learning_rate": 0.0002735652173913043,
946
+ "loss": 1.0427,
947
+ "step": 1470
948
+ },
949
+ {
950
+ "epoch": 0.47,
951
+ "learning_rate": 0.0002733719806763285,
952
+ "loss": 1.0333,
953
+ "step": 1480
954
+ },
955
+ {
956
+ "epoch": 0.48,
957
+ "learning_rate": 0.00027317874396135265,
958
+ "loss": 1.0084,
959
+ "step": 1490
960
+ },
961
+ {
962
+ "epoch": 0.48,
963
+ "learning_rate": 0.0002729855072463768,
964
+ "loss": 1.0555,
965
+ "step": 1500
966
+ },
967
+ {
968
+ "epoch": 0.48,
969
+ "learning_rate": 0.00027279227053140094,
970
+ "loss": 1.0419,
971
+ "step": 1510
972
+ },
973
+ {
974
+ "epoch": 0.49,
975
+ "learning_rate": 0.0002725990338164251,
976
+ "loss": 1.0846,
977
+ "step": 1520
978
+ },
979
+ {
980
+ "epoch": 0.49,
981
+ "learning_rate": 0.0002724057971014493,
982
+ "loss": 1.0375,
983
+ "step": 1530
984
+ },
985
+ {
986
+ "epoch": 0.49,
987
+ "learning_rate": 0.0002722125603864734,
988
+ "loss": 1.0593,
989
+ "step": 1540
990
+ },
991
+ {
992
+ "epoch": 0.5,
993
+ "learning_rate": 0.00027201932367149756,
994
+ "loss": 1.017,
995
+ "step": 1550
996
+ },
997
+ {
998
+ "epoch": 0.5,
999
+ "learning_rate": 0.0002718260869565217,
1000
+ "loss": 1.0887,
1001
+ "step": 1560
1002
+ },
1003
+ {
1004
+ "epoch": 0.5,
1005
+ "learning_rate": 0.0002716328502415459,
1006
+ "loss": 1.0333,
1007
+ "step": 1570
1008
+ },
1009
+ {
1010
+ "epoch": 0.51,
1011
+ "learning_rate": 0.00027143961352657004,
1012
+ "loss": 1.0439,
1013
+ "step": 1580
1014
+ },
1015
+ {
1016
+ "epoch": 0.51,
1017
+ "learning_rate": 0.0002712463768115942,
1018
+ "loss": 1.0221,
1019
+ "step": 1590
1020
+ },
1021
+ {
1022
+ "epoch": 0.51,
1023
+ "learning_rate": 0.00027105314009661833,
1024
+ "loss": 1.063,
1025
+ "step": 1600
1026
+ },
1027
+ {
1028
+ "epoch": 0.51,
1029
+ "eval_loss": 1.042041301727295,
1030
+ "eval_runtime": 196.9233,
1031
+ "eval_samples_per_second": 10.156,
1032
+ "eval_steps_per_second": 1.27,
1033
+ "step": 1600
1034
+ },
1035
+ {
1036
+ "epoch": 0.52,
1037
+ "learning_rate": 0.00027085990338164247,
1038
+ "loss": 1.0347,
1039
+ "step": 1610
1040
+ },
1041
+ {
1042
+ "epoch": 0.52,
1043
+ "learning_rate": 0.00027066666666666667,
1044
+ "loss": 1.027,
1045
+ "step": 1620
1046
+ },
1047
+ {
1048
+ "epoch": 0.52,
1049
+ "learning_rate": 0.0002704734299516908,
1050
+ "loss": 0.9961,
1051
+ "step": 1630
1052
+ },
1053
+ {
1054
+ "epoch": 0.52,
1055
+ "learning_rate": 0.00027028019323671495,
1056
+ "loss": 0.9966,
1057
+ "step": 1640
1058
+ },
1059
+ {
1060
+ "epoch": 0.53,
1061
+ "learning_rate": 0.0002700869565217391,
1062
+ "loss": 1.0474,
1063
+ "step": 1650
1064
+ },
1065
+ {
1066
+ "epoch": 0.53,
1067
+ "learning_rate": 0.0002698937198067633,
1068
+ "loss": 1.017,
1069
+ "step": 1660
1070
+ },
1071
+ {
1072
+ "epoch": 0.53,
1073
+ "learning_rate": 0.00026970048309178744,
1074
+ "loss": 1.0663,
1075
+ "step": 1670
1076
+ },
1077
+ {
1078
+ "epoch": 0.54,
1079
+ "learning_rate": 0.0002695072463768116,
1080
+ "loss": 1.0505,
1081
+ "step": 1680
1082
+ },
1083
+ {
1084
+ "epoch": 0.54,
1085
+ "learning_rate": 0.0002693140096618357,
1086
+ "loss": 1.0954,
1087
+ "step": 1690
1088
+ },
1089
+ {
1090
+ "epoch": 0.54,
1091
+ "learning_rate": 0.00026912077294685986,
1092
+ "loss": 1.0136,
1093
+ "step": 1700
1094
+ },
1095
+ {
1096
+ "epoch": 0.55,
1097
+ "learning_rate": 0.00026892753623188406,
1098
+ "loss": 1.0749,
1099
+ "step": 1710
1100
+ },
1101
+ {
1102
+ "epoch": 0.55,
1103
+ "learning_rate": 0.0002687342995169082,
1104
+ "loss": 0.9969,
1105
+ "step": 1720
1106
+ },
1107
+ {
1108
+ "epoch": 0.55,
1109
+ "learning_rate": 0.00026854106280193234,
1110
+ "loss": 1.0681,
1111
+ "step": 1730
1112
+ },
1113
+ {
1114
+ "epoch": 0.56,
1115
+ "learning_rate": 0.0002683478260869565,
1116
+ "loss": 1.0237,
1117
+ "step": 1740
1118
+ },
1119
+ {
1120
+ "epoch": 0.56,
1121
+ "learning_rate": 0.0002681545893719807,
1122
+ "loss": 1.0473,
1123
+ "step": 1750
1124
+ },
1125
+ {
1126
+ "epoch": 0.56,
1127
+ "learning_rate": 0.0002679613526570048,
1128
+ "loss": 1.0249,
1129
+ "step": 1760
1130
+ },
1131
+ {
1132
+ "epoch": 0.57,
1133
+ "learning_rate": 0.00026776811594202897,
1134
+ "loss": 0.999,
1135
+ "step": 1770
1136
+ },
1137
+ {
1138
+ "epoch": 0.57,
1139
+ "learning_rate": 0.0002675748792270531,
1140
+ "loss": 1.0299,
1141
+ "step": 1780
1142
+ },
1143
+ {
1144
+ "epoch": 0.57,
1145
+ "learning_rate": 0.0002673816425120773,
1146
+ "loss": 1.0885,
1147
+ "step": 1790
1148
+ },
1149
+ {
1150
+ "epoch": 0.58,
1151
+ "learning_rate": 0.00026718840579710145,
1152
+ "loss": 1.0552,
1153
+ "step": 1800
1154
+ },
1155
+ {
1156
+ "epoch": 0.58,
1157
+ "eval_loss": 1.0400397777557373,
1158
+ "eval_runtime": 196.3928,
1159
+ "eval_samples_per_second": 10.184,
1160
+ "eval_steps_per_second": 1.273,
1161
+ "step": 1800
1162
+ },
1163
+ {
1164
+ "epoch": 0.58,
1165
+ "learning_rate": 0.0002669951690821256,
1166
+ "loss": 1.0717,
1167
+ "step": 1810
1168
+ },
1169
+ {
1170
+ "epoch": 0.58,
1171
+ "learning_rate": 0.00026680193236714974,
1172
+ "loss": 1.0412,
1173
+ "step": 1820
1174
+ },
1175
+ {
1176
+ "epoch": 0.59,
1177
+ "learning_rate": 0.0002666086956521739,
1178
+ "loss": 1.0221,
1179
+ "step": 1830
1180
+ },
1181
+ {
1182
+ "epoch": 0.59,
1183
+ "learning_rate": 0.0002664154589371981,
1184
+ "loss": 1.0729,
1185
+ "step": 1840
1186
+ },
1187
+ {
1188
+ "epoch": 0.59,
1189
+ "learning_rate": 0.0002662222222222222,
1190
+ "loss": 1.0284,
1191
+ "step": 1850
1192
+ },
1193
+ {
1194
+ "epoch": 0.6,
1195
+ "learning_rate": 0.00026602898550724636,
1196
+ "loss": 1.0319,
1197
+ "step": 1860
1198
+ },
1199
+ {
1200
+ "epoch": 0.6,
1201
+ "learning_rate": 0.0002658357487922705,
1202
+ "loss": 1.0305,
1203
+ "step": 1870
1204
+ },
1205
+ {
1206
+ "epoch": 0.6,
1207
+ "learning_rate": 0.0002656425120772947,
1208
+ "loss": 1.0398,
1209
+ "step": 1880
1210
+ },
1211
+ {
1212
+ "epoch": 0.6,
1213
+ "learning_rate": 0.00026544927536231884,
1214
+ "loss": 1.0427,
1215
+ "step": 1890
1216
+ },
1217
+ {
1218
+ "epoch": 0.61,
1219
+ "learning_rate": 0.000265256038647343,
1220
+ "loss": 0.9895,
1221
+ "step": 1900
1222
+ },
1223
+ {
1224
+ "epoch": 0.61,
1225
+ "learning_rate": 0.00026506280193236713,
1226
+ "loss": 1.0631,
1227
+ "step": 1910
1228
+ },
1229
+ {
1230
+ "epoch": 0.61,
1231
+ "learning_rate": 0.00026486956521739127,
1232
+ "loss": 1.0165,
1233
+ "step": 1920
1234
+ },
1235
+ {
1236
+ "epoch": 0.62,
1237
+ "learning_rate": 0.00026467632850241547,
1238
+ "loss": 1.0257,
1239
+ "step": 1930
1240
+ },
1241
+ {
1242
+ "epoch": 0.62,
1243
+ "learning_rate": 0.0002644830917874396,
1244
+ "loss": 1.0475,
1245
+ "step": 1940
1246
+ },
1247
+ {
1248
+ "epoch": 0.62,
1249
+ "learning_rate": 0.00026428985507246375,
1250
+ "loss": 1.0641,
1251
+ "step": 1950
1252
+ },
1253
+ {
1254
+ "epoch": 0.63,
1255
+ "learning_rate": 0.0002640966183574879,
1256
+ "loss": 1.014,
1257
+ "step": 1960
1258
+ },
1259
+ {
1260
+ "epoch": 0.63,
1261
+ "learning_rate": 0.0002639033816425121,
1262
+ "loss": 1.021,
1263
+ "step": 1970
1264
+ },
1265
+ {
1266
+ "epoch": 0.63,
1267
+ "learning_rate": 0.00026371014492753623,
1268
+ "loss": 1.0223,
1269
+ "step": 1980
1270
+ },
1271
+ {
1272
+ "epoch": 0.64,
1273
+ "learning_rate": 0.0002635169082125604,
1274
+ "loss": 1.0313,
1275
+ "step": 1990
1276
+ },
1277
+ {
1278
+ "epoch": 0.64,
1279
+ "learning_rate": 0.0002633236714975845,
1280
+ "loss": 1.069,
1281
+ "step": 2000
1282
+ },
1283
+ {
1284
+ "epoch": 0.64,
1285
+ "eval_loss": 1.037448525428772,
1286
+ "eval_runtime": 196.1715,
1287
+ "eval_samples_per_second": 10.195,
1288
+ "eval_steps_per_second": 1.274,
1289
+ "step": 2000
1290
+ },
1291
+ {
1292
+ "epoch": 0.64,
1293
+ "learning_rate": 0.00026313043478260866,
1294
+ "loss": 1.0353,
1295
+ "step": 2010
1296
+ },
1297
+ {
1298
+ "epoch": 0.65,
1299
+ "learning_rate": 0.00026293719806763286,
1300
+ "loss": 1.0634,
1301
+ "step": 2020
1302
+ },
1303
+ {
1304
+ "epoch": 0.65,
1305
+ "learning_rate": 0.000262743961352657,
1306
+ "loss": 1.0551,
1307
+ "step": 2030
1308
+ },
1309
+ {
1310
+ "epoch": 0.65,
1311
+ "learning_rate": 0.00026255072463768114,
1312
+ "loss": 1.0555,
1313
+ "step": 2040
1314
+ },
1315
+ {
1316
+ "epoch": 0.66,
1317
+ "learning_rate": 0.0002623574879227053,
1318
+ "loss": 1.019,
1319
+ "step": 2050
1320
+ },
1321
+ {
1322
+ "epoch": 0.66,
1323
+ "learning_rate": 0.0002621642512077295,
1324
+ "loss": 1.0524,
1325
+ "step": 2060
1326
+ },
1327
+ {
1328
+ "epoch": 0.66,
1329
+ "learning_rate": 0.0002619710144927536,
1330
+ "loss": 1.0285,
1331
+ "step": 2070
1332
+ },
1333
+ {
1334
+ "epoch": 0.67,
1335
+ "learning_rate": 0.00026177777777777777,
1336
+ "loss": 1.0271,
1337
+ "step": 2080
1338
+ },
1339
+ {
1340
+ "epoch": 0.67,
1341
+ "learning_rate": 0.0002615845410628019,
1342
+ "loss": 1.0253,
1343
+ "step": 2090
1344
+ },
1345
+ {
1346
+ "epoch": 0.67,
1347
+ "learning_rate": 0.0002613913043478261,
1348
+ "loss": 1.0134,
1349
+ "step": 2100
1350
+ },
1351
+ {
1352
+ "epoch": 0.68,
1353
+ "learning_rate": 0.00026119806763285025,
1354
+ "loss": 1.0424,
1355
+ "step": 2110
1356
+ },
1357
+ {
1358
+ "epoch": 0.68,
1359
+ "learning_rate": 0.0002610048309178744,
1360
+ "loss": 1.0769,
1361
+ "step": 2120
1362
+ },
1363
+ {
1364
+ "epoch": 0.68,
1365
+ "learning_rate": 0.00026081159420289854,
1366
+ "loss": 1.0464,
1367
+ "step": 2130
1368
+ },
1369
+ {
1370
+ "epoch": 0.68,
1371
+ "learning_rate": 0.0002606183574879227,
1372
+ "loss": 1.0213,
1373
+ "step": 2140
1374
+ },
1375
+ {
1376
+ "epoch": 0.69,
1377
+ "learning_rate": 0.0002604251207729469,
1378
+ "loss": 1.0178,
1379
+ "step": 2150
1380
+ },
1381
+ {
1382
+ "epoch": 0.69,
1383
+ "learning_rate": 0.000260231884057971,
1384
+ "loss": 1.0113,
1385
+ "step": 2160
1386
+ },
1387
+ {
1388
+ "epoch": 0.69,
1389
+ "learning_rate": 0.00026003864734299516,
1390
+ "loss": 1.0261,
1391
+ "step": 2170
1392
+ },
1393
+ {
1394
+ "epoch": 0.7,
1395
+ "learning_rate": 0.0002598454106280193,
1396
+ "loss": 1.0522,
1397
+ "step": 2180
1398
+ },
1399
+ {
1400
+ "epoch": 0.7,
1401
+ "learning_rate": 0.00025965217391304344,
1402
+ "loss": 1.0553,
1403
+ "step": 2190
1404
+ },
1405
+ {
1406
+ "epoch": 0.7,
1407
+ "learning_rate": 0.00025945893719806764,
1408
+ "loss": 1.036,
1409
+ "step": 2200
1410
+ },
1411
+ {
1412
+ "epoch": 0.7,
1413
+ "eval_loss": 1.0350252389907837,
1414
+ "eval_runtime": 195.1168,
1415
+ "eval_samples_per_second": 10.25,
1416
+ "eval_steps_per_second": 1.281,
1417
+ "step": 2200
1418
+ },
1419
+ {
1420
+ "epoch": 0.71,
1421
+ "learning_rate": 0.0002592657004830918,
1422
+ "loss": 1.0341,
1423
+ "step": 2210
1424
+ },
1425
+ {
1426
+ "epoch": 0.71,
1427
+ "learning_rate": 0.0002590724637681159,
1428
+ "loss": 1.0552,
1429
+ "step": 2220
1430
+ },
1431
+ {
1432
+ "epoch": 0.71,
1433
+ "learning_rate": 0.00025887922705314007,
1434
+ "loss": 0.9925,
1435
+ "step": 2230
1436
+ },
1437
+ {
1438
+ "epoch": 0.72,
1439
+ "learning_rate": 0.00025868599033816427,
1440
+ "loss": 1.0427,
1441
+ "step": 2240
1442
+ },
1443
+ {
1444
+ "epoch": 0.72,
1445
+ "learning_rate": 0.0002584927536231884,
1446
+ "loss": 1.0169,
1447
+ "step": 2250
1448
+ },
1449
+ {
1450
+ "epoch": 0.72,
1451
+ "learning_rate": 0.00025829951690821255,
1452
+ "loss": 1.028,
1453
+ "step": 2260
1454
+ },
1455
+ {
1456
+ "epoch": 0.73,
1457
+ "learning_rate": 0.0002581062801932367,
1458
+ "loss": 1.0556,
1459
+ "step": 2270
1460
+ },
1461
+ {
1462
+ "epoch": 0.73,
1463
+ "learning_rate": 0.00025791304347826084,
1464
+ "loss": 1.0242,
1465
+ "step": 2280
1466
+ },
1467
+ {
1468
+ "epoch": 0.73,
1469
+ "learning_rate": 0.00025771980676328503,
1470
+ "loss": 1.0307,
1471
+ "step": 2290
1472
+ },
1473
+ {
1474
+ "epoch": 0.74,
1475
+ "learning_rate": 0.0002575265700483092,
1476
+ "loss": 1.0083,
1477
+ "step": 2300
1478
+ },
1479
+ {
1480
+ "epoch": 0.74,
1481
+ "learning_rate": 0.0002573333333333333,
1482
+ "loss": 1.0146,
1483
+ "step": 2310
1484
+ },
1485
+ {
1486
+ "epoch": 0.74,
1487
+ "learning_rate": 0.00025714009661835746,
1488
+ "loss": 0.993,
1489
+ "step": 2320
1490
+ },
1491
+ {
1492
+ "epoch": 0.75,
1493
+ "learning_rate": 0.0002569468599033816,
1494
+ "loss": 1.0726,
1495
+ "step": 2330
1496
+ },
1497
+ {
1498
+ "epoch": 0.75,
1499
+ "learning_rate": 0.0002567536231884058,
1500
+ "loss": 1.028,
1501
+ "step": 2340
1502
+ },
1503
+ {
1504
+ "epoch": 0.75,
1505
+ "learning_rate": 0.00025656038647342994,
1506
+ "loss": 1.031,
1507
+ "step": 2350
1508
+ },
1509
+ {
1510
+ "epoch": 0.76,
1511
+ "learning_rate": 0.0002563671497584541,
1512
+ "loss": 1.0175,
1513
+ "step": 2360
1514
+ },
1515
+ {
1516
+ "epoch": 0.76,
1517
+ "learning_rate": 0.00025617391304347823,
1518
+ "loss": 1.0618,
1519
+ "step": 2370
1520
+ },
1521
+ {
1522
+ "epoch": 0.76,
1523
+ "learning_rate": 0.0002559806763285024,
1524
+ "loss": 1.0442,
1525
+ "step": 2380
1526
+ },
1527
+ {
1528
+ "epoch": 0.76,
1529
+ "learning_rate": 0.00025578743961352657,
1530
+ "loss": 1.0507,
1531
+ "step": 2390
1532
+ },
1533
+ {
1534
+ "epoch": 0.77,
1535
+ "learning_rate": 0.0002555942028985507,
1536
+ "loss": 1.0347,
1537
+ "step": 2400
1538
+ },
1539
+ {
1540
+ "epoch": 0.77,
1541
+ "eval_loss": 1.0352022647857666,
1542
+ "eval_runtime": 194.821,
1543
+ "eval_samples_per_second": 10.266,
1544
+ "eval_steps_per_second": 1.283,
1545
+ "step": 2400
1546
+ },
1547
+ {
1548
+ "epoch": 0.77,
1549
+ "learning_rate": 0.00025540096618357485,
1550
+ "loss": 1.0786,
1551
+ "step": 2410
1552
+ },
1553
+ {
1554
+ "epoch": 0.77,
1555
+ "learning_rate": 0.000255207729468599,
1556
+ "loss": 1.0542,
1557
+ "step": 2420
1558
+ },
1559
+ {
1560
+ "epoch": 0.78,
1561
+ "learning_rate": 0.0002550144927536232,
1562
+ "loss": 1.0799,
1563
+ "step": 2430
1564
+ },
1565
+ {
1566
+ "epoch": 0.78,
1567
+ "learning_rate": 0.00025482125603864733,
1568
+ "loss": 1.0518,
1569
+ "step": 2440
1570
+ },
1571
+ {
1572
+ "epoch": 0.78,
1573
+ "learning_rate": 0.0002546280193236715,
1574
+ "loss": 1.005,
1575
+ "step": 2450
1576
+ },
1577
+ {
1578
+ "epoch": 0.79,
1579
+ "learning_rate": 0.0002544347826086956,
1580
+ "loss": 1.0239,
1581
+ "step": 2460
1582
+ },
1583
+ {
1584
+ "epoch": 0.79,
1585
+ "learning_rate": 0.0002542415458937198,
1586
+ "loss": 1.0243,
1587
+ "step": 2470
1588
+ },
1589
+ {
1590
+ "epoch": 0.79,
1591
+ "learning_rate": 0.00025404830917874396,
1592
+ "loss": 1.0144,
1593
+ "step": 2480
1594
+ },
1595
+ {
1596
+ "epoch": 0.8,
1597
+ "learning_rate": 0.0002538550724637681,
1598
+ "loss": 0.9995,
1599
+ "step": 2490
1600
+ },
1601
+ {
1602
+ "epoch": 0.8,
1603
+ "learning_rate": 0.00025366183574879224,
1604
+ "loss": 1.0222,
1605
+ "step": 2500
1606
+ },
1607
+ {
1608
+ "epoch": 0.8,
1609
+ "learning_rate": 0.0002534685990338164,
1610
+ "loss": 1.0381,
1611
+ "step": 2510
1612
+ },
1613
+ {
1614
+ "epoch": 0.81,
1615
+ "learning_rate": 0.0002532753623188406,
1616
+ "loss": 1.0309,
1617
+ "step": 2520
1618
+ },
1619
+ {
1620
+ "epoch": 0.81,
1621
+ "learning_rate": 0.0002530821256038647,
1622
+ "loss": 1.0176,
1623
+ "step": 2530
1624
+ },
1625
+ {
1626
+ "epoch": 0.81,
1627
+ "learning_rate": 0.00025288888888888887,
1628
+ "loss": 1.0085,
1629
+ "step": 2540
1630
+ },
1631
+ {
1632
+ "epoch": 0.82,
1633
+ "learning_rate": 0.000252695652173913,
1634
+ "loss": 1.049,
1635
+ "step": 2550
1636
+ },
1637
+ {
1638
+ "epoch": 0.82,
1639
+ "learning_rate": 0.00025250241545893715,
1640
+ "loss": 1.0146,
1641
+ "step": 2560
1642
+ },
1643
+ {
1644
+ "epoch": 0.82,
1645
+ "learning_rate": 0.00025230917874396135,
1646
+ "loss": 1.0584,
1647
+ "step": 2570
1648
+ },
1649
+ {
1650
+ "epoch": 0.83,
1651
+ "learning_rate": 0.0002521159420289855,
1652
+ "loss": 1.0562,
1653
+ "step": 2580
1654
+ },
1655
+ {
1656
+ "epoch": 0.83,
1657
+ "learning_rate": 0.00025192270531400963,
1658
+ "loss": 1.0391,
1659
+ "step": 2590
1660
+ },
1661
+ {
1662
+ "epoch": 0.83,
1663
+ "learning_rate": 0.0002517294685990338,
1664
+ "loss": 1.0388,
1665
+ "step": 2600
1666
+ },
1667
+ {
1668
+ "epoch": 0.83,
1669
+ "eval_loss": 1.0332427024841309,
1670
+ "eval_runtime": 194.4995,
1671
+ "eval_samples_per_second": 10.283,
1672
+ "eval_steps_per_second": 1.285,
1673
+ "step": 2600
1674
+ },
1675
+ {
1676
+ "epoch": 0.84,
1677
+ "learning_rate": 0.000251536231884058,
1678
+ "loss": 1.0589,
1679
+ "step": 2610
1680
+ },
1681
+ {
1682
+ "epoch": 0.84,
1683
+ "learning_rate": 0.0002513429951690821,
1684
+ "loss": 1.0123,
1685
+ "step": 2620
1686
+ },
1687
+ {
1688
+ "epoch": 0.84,
1689
+ "learning_rate": 0.00025114975845410626,
1690
+ "loss": 0.9994,
1691
+ "step": 2630
1692
+ },
1693
+ {
1694
+ "epoch": 0.84,
1695
+ "learning_rate": 0.0002509565217391304,
1696
+ "loss": 1.0468,
1697
+ "step": 2640
1698
+ },
1699
+ {
1700
+ "epoch": 0.85,
1701
+ "learning_rate": 0.00025076328502415454,
1702
+ "loss": 1.0455,
1703
+ "step": 2650
1704
+ },
1705
+ {
1706
+ "epoch": 0.85,
1707
+ "learning_rate": 0.00025057004830917874,
1708
+ "loss": 1.0127,
1709
+ "step": 2660
1710
+ },
1711
+ {
1712
+ "epoch": 0.85,
1713
+ "learning_rate": 0.0002503768115942029,
1714
+ "loss": 1.0031,
1715
+ "step": 2670
1716
+ },
1717
+ {
1718
+ "epoch": 0.86,
1719
+ "learning_rate": 0.000250183574879227,
1720
+ "loss": 1.0568,
1721
+ "step": 2680
1722
+ },
1723
+ {
1724
+ "epoch": 0.86,
1725
+ "learning_rate": 0.00024999033816425117,
1726
+ "loss": 1.0687,
1727
+ "step": 2690
1728
+ },
1729
+ {
1730
+ "epoch": 0.86,
1731
+ "learning_rate": 0.00024979710144927537,
1732
+ "loss": 1.0378,
1733
+ "step": 2700
1734
+ },
1735
+ {
1736
+ "epoch": 0.87,
1737
+ "learning_rate": 0.0002496038647342995,
1738
+ "loss": 1.0761,
1739
+ "step": 2710
1740
+ },
1741
+ {
1742
+ "epoch": 0.87,
1743
+ "learning_rate": 0.00024941062801932365,
1744
+ "loss": 0.9987,
1745
+ "step": 2720
1746
+ },
1747
+ {
1748
+ "epoch": 0.87,
1749
+ "learning_rate": 0.0002492173913043478,
1750
+ "loss": 1.0179,
1751
+ "step": 2730
1752
+ },
1753
+ {
1754
+ "epoch": 0.88,
1755
+ "learning_rate": 0.00024902415458937194,
1756
+ "loss": 1.0491,
1757
+ "step": 2740
1758
+ },
1759
+ {
1760
+ "epoch": 0.88,
1761
+ "learning_rate": 0.00024883091787439613,
1762
+ "loss": 1.0199,
1763
+ "step": 2750
1764
+ },
1765
+ {
1766
+ "epoch": 0.88,
1767
+ "learning_rate": 0.0002486376811594203,
1768
+ "loss": 1.062,
1769
+ "step": 2760
1770
+ },
1771
+ {
1772
+ "epoch": 0.89,
1773
+ "learning_rate": 0.0002484444444444444,
1774
+ "loss": 1.0389,
1775
+ "step": 2770
1776
+ },
1777
+ {
1778
+ "epoch": 0.89,
1779
+ "learning_rate": 0.00024825120772946856,
1780
+ "loss": 1.027,
1781
+ "step": 2780
1782
+ },
1783
+ {
1784
+ "epoch": 0.89,
1785
+ "learning_rate": 0.0002480579710144927,
1786
+ "loss": 1.0093,
1787
+ "step": 2790
1788
+ },
1789
+ {
1790
+ "epoch": 0.9,
1791
+ "learning_rate": 0.0002478647342995169,
1792
+ "loss": 1.0448,
1793
+ "step": 2800
1794
+ },
1795
+ {
1796
+ "epoch": 0.9,
1797
+ "eval_loss": 1.032317876815796,
1798
+ "eval_runtime": 196.407,
1799
+ "eval_samples_per_second": 10.183,
1800
+ "eval_steps_per_second": 1.273,
1801
+ "step": 2800
1802
+ },
1803
+ {
1804
+ "epoch": 0.9,
1805
+ "learning_rate": 0.00024767149758454104,
1806
+ "loss": 1.0627,
1807
+ "step": 2810
1808
+ },
1809
+ {
1810
+ "epoch": 0.9,
1811
+ "learning_rate": 0.0002474782608695652,
1812
+ "loss": 1.0202,
1813
+ "step": 2820
1814
+ },
1815
+ {
1816
+ "epoch": 0.91,
1817
+ "learning_rate": 0.00024728502415458933,
1818
+ "loss": 1.0905,
1819
+ "step": 2830
1820
+ },
1821
+ {
1822
+ "epoch": 0.91,
1823
+ "learning_rate": 0.0002470917874396135,
1824
+ "loss": 1.0552,
1825
+ "step": 2840
1826
+ },
1827
+ {
1828
+ "epoch": 0.91,
1829
+ "learning_rate": 0.00024689855072463767,
1830
+ "loss": 1.0605,
1831
+ "step": 2850
1832
+ },
1833
+ {
1834
+ "epoch": 0.92,
1835
+ "learning_rate": 0.0002467053140096618,
1836
+ "loss": 1.0194,
1837
+ "step": 2860
1838
+ },
1839
+ {
1840
+ "epoch": 0.92,
1841
+ "learning_rate": 0.00024651207729468595,
1842
+ "loss": 1.077,
1843
+ "step": 2870
1844
+ },
1845
+ {
1846
+ "epoch": 0.92,
1847
+ "learning_rate": 0.0002463188405797101,
1848
+ "loss": 1.0845,
1849
+ "step": 2880
1850
+ },
1851
+ {
1852
+ "epoch": 0.92,
1853
+ "learning_rate": 0.0002461256038647343,
1854
+ "loss": 1.0225,
1855
+ "step": 2890
1856
+ },
1857
+ {
1858
+ "epoch": 0.93,
1859
+ "learning_rate": 0.00024593236714975843,
1860
+ "loss": 1.0217,
1861
+ "step": 2900
1862
+ },
1863
+ {
1864
+ "epoch": 0.93,
1865
+ "learning_rate": 0.0002457391304347826,
1866
+ "loss": 1.0415,
1867
+ "step": 2910
1868
+ },
1869
+ {
1870
+ "epoch": 0.93,
1871
+ "learning_rate": 0.0002455458937198067,
1872
+ "loss": 1.0581,
1873
+ "step": 2920
1874
+ },
1875
+ {
1876
+ "epoch": 0.94,
1877
+ "learning_rate": 0.00024535265700483086,
1878
+ "loss": 1.0459,
1879
+ "step": 2930
1880
+ },
1881
+ {
1882
+ "epoch": 0.94,
1883
+ "learning_rate": 0.00024515942028985506,
1884
+ "loss": 1.0505,
1885
+ "step": 2940
1886
+ },
1887
+ {
1888
+ "epoch": 0.94,
1889
+ "learning_rate": 0.0002449661835748792,
1890
+ "loss": 1.0562,
1891
+ "step": 2950
1892
+ },
1893
+ {
1894
+ "epoch": 0.95,
1895
+ "learning_rate": 0.00024477294685990334,
1896
+ "loss": 0.9631,
1897
+ "step": 2960
1898
+ },
1899
+ {
1900
+ "epoch": 0.95,
1901
+ "learning_rate": 0.0002445797101449275,
1902
+ "loss": 1.0287,
1903
+ "step": 2970
1904
+ },
1905
+ {
1906
+ "epoch": 0.95,
1907
+ "learning_rate": 0.0002443864734299517,
1908
+ "loss": 1.0227,
1909
+ "step": 2980
1910
+ },
1911
+ {
1912
+ "epoch": 0.96,
1913
+ "learning_rate": 0.0002441932367149758,
1914
+ "loss": 1.03,
1915
+ "step": 2990
1916
+ },
1917
+ {
1918
+ "epoch": 0.96,
1919
+ "learning_rate": 0.000244,
1920
+ "loss": 1.0795,
1921
+ "step": 3000
1922
+ },
1923
+ {
1924
+ "epoch": 0.96,
1925
+ "eval_loss": 1.0316544771194458,
1926
+ "eval_runtime": 197.1671,
1927
+ "eval_samples_per_second": 10.144,
1928
+ "eval_steps_per_second": 1.268,
1929
+ "step": 3000
1930
+ }
1931
+ ],
1932
+ "logging_steps": 10,
1933
+ "max_steps": 15625,
1934
+ "num_train_epochs": 5,
1935
+ "save_steps": 200,
1936
+ "total_flos": 6.448770827226317e+17,
1937
+ "trial_name": null,
1938
+ "trial_params": null
1939
+ }
text_models/llama-13b/best_model/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:38c31c7207727a5321d909f045304236d62a16c24dbbfbf145c5b58dc58f0b3a
3
+ size 4027