ryusangwon commited on
Commit
fa11b45
1 Parent(s): 1428b6b

Model save

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. 119_Llama-2-13b-hf/README.md +55 -0
  2. 119_Llama-2-13b-hf/adapter_config.json +21 -0
  3. 119_Llama-2-13b-hf/adapter_model.safetensors +3 -0
  4. 119_Llama-2-13b-hf/checkpoint-1000/README.md +9 -0
  5. 119_Llama-2-13b-hf/checkpoint-1000/adapter_config.json +21 -0
  6. 119_Llama-2-13b-hf/checkpoint-1000/adapter_model.safetensors +3 -0
  7. 119_Llama-2-13b-hf/checkpoint-1000/optimizer.pt +3 -0
  8. 119_Llama-2-13b-hf/checkpoint-1000/rng_state.pth +3 -0
  9. 119_Llama-2-13b-hf/checkpoint-1000/scheduler.pt +3 -0
  10. 119_Llama-2-13b-hf/checkpoint-1000/trainer_state.json +621 -0
  11. 119_Llama-2-13b-hf/checkpoint-1000/training_args.bin +3 -0
  12. 119_Llama-2-13b-hf/checkpoint-1500/README.md +9 -0
  13. 119_Llama-2-13b-hf/checkpoint-1500/adapter_config.json +21 -0
  14. 119_Llama-2-13b-hf/checkpoint-1500/adapter_model.safetensors +3 -0
  15. 119_Llama-2-13b-hf/checkpoint-1500/optimizer.pt +3 -0
  16. 119_Llama-2-13b-hf/checkpoint-1500/rng_state.pth +3 -0
  17. 119_Llama-2-13b-hf/checkpoint-1500/scheduler.pt +3 -0
  18. 119_Llama-2-13b-hf/checkpoint-1500/trainer_state.json +921 -0
  19. 119_Llama-2-13b-hf/checkpoint-1500/training_args.bin +3 -0
  20. 119_Llama-2-13b-hf/checkpoint-2000/README.md +9 -0
  21. 119_Llama-2-13b-hf/checkpoint-2000/adapter_config.json +21 -0
  22. 119_Llama-2-13b-hf/checkpoint-2000/adapter_model.safetensors +3 -0
  23. 119_Llama-2-13b-hf/checkpoint-2000/optimizer.pt +3 -0
  24. 119_Llama-2-13b-hf/checkpoint-2000/rng_state.pth +3 -0
  25. 119_Llama-2-13b-hf/checkpoint-2000/scheduler.pt +3 -0
  26. 119_Llama-2-13b-hf/checkpoint-2000/trainer_state.json +1221 -0
  27. 119_Llama-2-13b-hf/checkpoint-2000/training_args.bin +3 -0
  28. 119_Llama-2-13b-hf/checkpoint-2500/README.md +9 -0
  29. 119_Llama-2-13b-hf/checkpoint-2500/adapter_config.json +21 -0
  30. 119_Llama-2-13b-hf/checkpoint-2500/adapter_model.safetensors +3 -0
  31. 119_Llama-2-13b-hf/checkpoint-2500/optimizer.pt +3 -0
  32. 119_Llama-2-13b-hf/checkpoint-2500/rng_state.pth +3 -0
  33. 119_Llama-2-13b-hf/checkpoint-2500/scheduler.pt +3 -0
  34. 119_Llama-2-13b-hf/checkpoint-2500/trainer_state.json +1521 -0
  35. 119_Llama-2-13b-hf/checkpoint-2500/training_args.bin +3 -0
  36. 119_Llama-2-13b-hf/checkpoint-3000/README.md +9 -0
  37. 119_Llama-2-13b-hf/checkpoint-3000/adapter_config.json +21 -0
  38. 119_Llama-2-13b-hf/checkpoint-3000/adapter_model.safetensors +3 -0
  39. 119_Llama-2-13b-hf/checkpoint-3000/optimizer.pt +3 -0
  40. 119_Llama-2-13b-hf/checkpoint-3000/rng_state.pth +3 -0
  41. 119_Llama-2-13b-hf/checkpoint-3000/scheduler.pt +3 -0
  42. 119_Llama-2-13b-hf/checkpoint-3000/trainer_state.json +1821 -0
  43. 119_Llama-2-13b-hf/checkpoint-3000/training_args.bin +3 -0
  44. 119_Llama-2-13b-hf/checkpoint-3500/README.md +9 -0
  45. 119_Llama-2-13b-hf/checkpoint-3500/adapter_config.json +21 -0
  46. 119_Llama-2-13b-hf/checkpoint-3500/adapter_model.safetensors +3 -0
  47. 119_Llama-2-13b-hf/checkpoint-3500/optimizer.pt +3 -0
  48. 119_Llama-2-13b-hf/checkpoint-3500/rng_state.pth +3 -0
  49. 119_Llama-2-13b-hf/checkpoint-3500/scheduler.pt +3 -0
  50. 119_Llama-2-13b-hf/checkpoint-3500/trainer_state.json +2121 -0
119_Llama-2-13b-hf/README.md ADDED
@@ -0,0 +1,55 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: meta-llama/Llama-2-13b-hf
3
+ tags:
4
+ - generated_from_trainer
5
+ datasets:
6
+ - cnn_dailymail
7
+ model-index:
8
+ - name: 119_Llama-2-13b-hf
9
+ results: []
10
+ library_name: peft
11
+ ---
12
+
13
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
14
+ should probably proofread and complete it, then remove this comment. -->
15
+
16
+ # 119_Llama-2-13b-hf
17
+
18
+ This model is a fine-tuned version of [meta-llama/Llama-2-13b-hf](https://huggingface.co/meta-llama/Llama-2-13b-hf) on the cnn_dailymail dataset.
19
+
20
+ ## Model description
21
+
22
+ More information needed
23
+
24
+ ## Intended uses & limitations
25
+
26
+ More information needed
27
+
28
+ ## Training and evaluation data
29
+
30
+ More information needed
31
+
32
+ ## Training procedure
33
+
34
+ ### Training hyperparameters
35
+
36
+ The following hyperparameters were used during training:
37
+ - learning_rate: 5e-05
38
+ - train_batch_size: 4
39
+ - eval_batch_size: 4
40
+ - seed: 42
41
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
42
+ - lr_scheduler_type: linear
43
+ - num_epochs: 3
44
+
45
+ ### Training results
46
+
47
+
48
+
49
+ ### Framework versions
50
+
51
+ - PEFT 0.4.0
52
+ - Transformers 4.36.2
53
+ - Pytorch 2.0.1+cu117
54
+ - Datasets 2.15.0
55
+ - Tokenizers 0.15.0
119_Llama-2-13b-hf/adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "meta-llama/Llama-2-13b-hf",
4
+ "bias": "none",
5
+ "fan_in_fan_out": false,
6
+ "inference_mode": true,
7
+ "init_lora_weights": true,
8
+ "layers_pattern": null,
9
+ "layers_to_transform": null,
10
+ "lora_alpha": 32,
11
+ "lora_dropout": 0.05,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 8,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "q_proj",
18
+ "v_proj"
19
+ ],
20
+ "task_type": "CAUSAL_LM"
21
+ }
119_Llama-2-13b-hf/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9d039835e9adbd8744f799b20ca8941216f99cf8042362a3ea1055a0f8fced7a
3
+ size 26235704
119_Llama-2-13b-hf/checkpoint-1000/README.md ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+ ### Framework versions
7
+
8
+
9
+ - PEFT 0.4.0
119_Llama-2-13b-hf/checkpoint-1000/adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "meta-llama/Llama-2-13b-hf",
4
+ "bias": "none",
5
+ "fan_in_fan_out": false,
6
+ "inference_mode": true,
7
+ "init_lora_weights": true,
8
+ "layers_pattern": null,
9
+ "layers_to_transform": null,
10
+ "lora_alpha": 32,
11
+ "lora_dropout": 0.05,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 8,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "q_proj",
18
+ "v_proj"
19
+ ],
20
+ "task_type": "CAUSAL_LM"
21
+ }
119_Llama-2-13b-hf/checkpoint-1000/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2d0d46a8f3bbb68472179a2fc0fa551267aec0689253e1953701f7bd648450b1
3
+ size 26235704
119_Llama-2-13b-hf/checkpoint-1000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9ec34737103a4baed755c37eaf3407cd010d4b29dd81a5ca2c78b4c1b1a6cfc0
3
+ size 52567237
119_Llama-2-13b-hf/checkpoint-1000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2e81f118b452f4863e1ab9c13d95ca4f822c8af2f735ac6e8329b425adc3926b
3
+ size 14575
119_Llama-2-13b-hf/checkpoint-1000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:000b0908f78286a797ebcf8e6cca70fe5a57b3d73df134a8c09baa379fbdcb40
3
+ size 627
119_Llama-2-13b-hf/checkpoint-1000/trainer_state.json ADDED
@@ -0,0 +1,621 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.7468259895444361,
5
+ "eval_steps": 500,
6
+ "global_step": 1000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.01,
13
+ "learning_rate": 4.98755290017426e-05,
14
+ "loss": 1.7891,
15
+ "step": 10
16
+ },
17
+ {
18
+ "epoch": 0.01,
19
+ "learning_rate": 4.975105800348519e-05,
20
+ "loss": 1.7869,
21
+ "step": 20
22
+ },
23
+ {
24
+ "epoch": 0.02,
25
+ "learning_rate": 4.962658700522779e-05,
26
+ "loss": 1.7868,
27
+ "step": 30
28
+ },
29
+ {
30
+ "epoch": 0.03,
31
+ "learning_rate": 4.9502116006970375e-05,
32
+ "loss": 1.7547,
33
+ "step": 40
34
+ },
35
+ {
36
+ "epoch": 0.04,
37
+ "learning_rate": 4.937764500871298e-05,
38
+ "loss": 1.6772,
39
+ "step": 50
40
+ },
41
+ {
42
+ "epoch": 0.04,
43
+ "learning_rate": 4.9253174010455565e-05,
44
+ "loss": 1.7076,
45
+ "step": 60
46
+ },
47
+ {
48
+ "epoch": 0.05,
49
+ "learning_rate": 4.912870301219816e-05,
50
+ "loss": 1.6948,
51
+ "step": 70
52
+ },
53
+ {
54
+ "epoch": 0.06,
55
+ "learning_rate": 4.9004232013940755e-05,
56
+ "loss": 1.7135,
57
+ "step": 80
58
+ },
59
+ {
60
+ "epoch": 0.07,
61
+ "learning_rate": 4.887976101568335e-05,
62
+ "loss": 1.7007,
63
+ "step": 90
64
+ },
65
+ {
66
+ "epoch": 0.07,
67
+ "learning_rate": 4.8755290017425944e-05,
68
+ "loss": 1.7247,
69
+ "step": 100
70
+ },
71
+ {
72
+ "epoch": 0.08,
73
+ "learning_rate": 4.863081901916854e-05,
74
+ "loss": 1.719,
75
+ "step": 110
76
+ },
77
+ {
78
+ "epoch": 0.09,
79
+ "learning_rate": 4.850634802091113e-05,
80
+ "loss": 1.7209,
81
+ "step": 120
82
+ },
83
+ {
84
+ "epoch": 0.1,
85
+ "learning_rate": 4.838187702265373e-05,
86
+ "loss": 1.7354,
87
+ "step": 130
88
+ },
89
+ {
90
+ "epoch": 0.1,
91
+ "learning_rate": 4.825740602439632e-05,
92
+ "loss": 1.7025,
93
+ "step": 140
94
+ },
95
+ {
96
+ "epoch": 0.11,
97
+ "learning_rate": 4.813293502613891e-05,
98
+ "loss": 1.6664,
99
+ "step": 150
100
+ },
101
+ {
102
+ "epoch": 0.12,
103
+ "learning_rate": 4.800846402788151e-05,
104
+ "loss": 1.6851,
105
+ "step": 160
106
+ },
107
+ {
108
+ "epoch": 0.13,
109
+ "learning_rate": 4.78839930296241e-05,
110
+ "loss": 1.6749,
111
+ "step": 170
112
+ },
113
+ {
114
+ "epoch": 0.13,
115
+ "learning_rate": 4.7759522031366697e-05,
116
+ "loss": 1.6887,
117
+ "step": 180
118
+ },
119
+ {
120
+ "epoch": 0.14,
121
+ "learning_rate": 4.763505103310929e-05,
122
+ "loss": 1.6571,
123
+ "step": 190
124
+ },
125
+ {
126
+ "epoch": 0.15,
127
+ "learning_rate": 4.751058003485188e-05,
128
+ "loss": 1.7136,
129
+ "step": 200
130
+ },
131
+ {
132
+ "epoch": 0.16,
133
+ "learning_rate": 4.738610903659448e-05,
134
+ "loss": 1.7098,
135
+ "step": 210
136
+ },
137
+ {
138
+ "epoch": 0.16,
139
+ "learning_rate": 4.726163803833707e-05,
140
+ "loss": 1.7285,
141
+ "step": 220
142
+ },
143
+ {
144
+ "epoch": 0.17,
145
+ "learning_rate": 4.7137167040079664e-05,
146
+ "loss": 1.7038,
147
+ "step": 230
148
+ },
149
+ {
150
+ "epoch": 0.18,
151
+ "learning_rate": 4.701269604182226e-05,
152
+ "loss": 1.692,
153
+ "step": 240
154
+ },
155
+ {
156
+ "epoch": 0.19,
157
+ "learning_rate": 4.6888225043564854e-05,
158
+ "loss": 1.6825,
159
+ "step": 250
160
+ },
161
+ {
162
+ "epoch": 0.19,
163
+ "learning_rate": 4.676375404530744e-05,
164
+ "loss": 1.7254,
165
+ "step": 260
166
+ },
167
+ {
168
+ "epoch": 0.2,
169
+ "learning_rate": 4.6639283047050044e-05,
170
+ "loss": 1.6765,
171
+ "step": 270
172
+ },
173
+ {
174
+ "epoch": 0.21,
175
+ "learning_rate": 4.651481204879263e-05,
176
+ "loss": 1.6796,
177
+ "step": 280
178
+ },
179
+ {
180
+ "epoch": 0.22,
181
+ "learning_rate": 4.6390341050535227e-05,
182
+ "loss": 1.7302,
183
+ "step": 290
184
+ },
185
+ {
186
+ "epoch": 0.22,
187
+ "learning_rate": 4.626587005227782e-05,
188
+ "loss": 1.7469,
189
+ "step": 300
190
+ },
191
+ {
192
+ "epoch": 0.23,
193
+ "learning_rate": 4.6141399054020416e-05,
194
+ "loss": 1.6345,
195
+ "step": 310
196
+ },
197
+ {
198
+ "epoch": 0.24,
199
+ "learning_rate": 4.601692805576301e-05,
200
+ "loss": 1.6761,
201
+ "step": 320
202
+ },
203
+ {
204
+ "epoch": 0.25,
205
+ "learning_rate": 4.5892457057505606e-05,
206
+ "loss": 1.6602,
207
+ "step": 330
208
+ },
209
+ {
210
+ "epoch": 0.25,
211
+ "learning_rate": 4.5767986059248194e-05,
212
+ "loss": 1.6637,
213
+ "step": 340
214
+ },
215
+ {
216
+ "epoch": 0.26,
217
+ "learning_rate": 4.5643515060990796e-05,
218
+ "loss": 1.6953,
219
+ "step": 350
220
+ },
221
+ {
222
+ "epoch": 0.27,
223
+ "learning_rate": 4.5519044062733384e-05,
224
+ "loss": 1.6829,
225
+ "step": 360
226
+ },
227
+ {
228
+ "epoch": 0.28,
229
+ "learning_rate": 4.539457306447598e-05,
230
+ "loss": 1.657,
231
+ "step": 370
232
+ },
233
+ {
234
+ "epoch": 0.28,
235
+ "learning_rate": 4.5270102066218574e-05,
236
+ "loss": 1.647,
237
+ "step": 380
238
+ },
239
+ {
240
+ "epoch": 0.29,
241
+ "learning_rate": 4.514563106796117e-05,
242
+ "loss": 1.6854,
243
+ "step": 390
244
+ },
245
+ {
246
+ "epoch": 0.3,
247
+ "learning_rate": 4.502116006970376e-05,
248
+ "loss": 1.6737,
249
+ "step": 400
250
+ },
251
+ {
252
+ "epoch": 0.31,
253
+ "learning_rate": 4.489668907144636e-05,
254
+ "loss": 1.6938,
255
+ "step": 410
256
+ },
257
+ {
258
+ "epoch": 0.31,
259
+ "learning_rate": 4.4772218073188946e-05,
260
+ "loss": 1.6721,
261
+ "step": 420
262
+ },
263
+ {
264
+ "epoch": 0.32,
265
+ "learning_rate": 4.464774707493155e-05,
266
+ "loss": 1.6683,
267
+ "step": 430
268
+ },
269
+ {
270
+ "epoch": 0.33,
271
+ "learning_rate": 4.4523276076674136e-05,
272
+ "loss": 1.6229,
273
+ "step": 440
274
+ },
275
+ {
276
+ "epoch": 0.34,
277
+ "learning_rate": 4.439880507841673e-05,
278
+ "loss": 1.6593,
279
+ "step": 450
280
+ },
281
+ {
282
+ "epoch": 0.34,
283
+ "learning_rate": 4.4274334080159326e-05,
284
+ "loss": 1.6844,
285
+ "step": 460
286
+ },
287
+ {
288
+ "epoch": 0.35,
289
+ "learning_rate": 4.414986308190192e-05,
290
+ "loss": 1.6754,
291
+ "step": 470
292
+ },
293
+ {
294
+ "epoch": 0.36,
295
+ "learning_rate": 4.4025392083644516e-05,
296
+ "loss": 1.6747,
297
+ "step": 480
298
+ },
299
+ {
300
+ "epoch": 0.37,
301
+ "learning_rate": 4.390092108538711e-05,
302
+ "loss": 1.6789,
303
+ "step": 490
304
+ },
305
+ {
306
+ "epoch": 0.37,
307
+ "learning_rate": 4.37764500871297e-05,
308
+ "loss": 1.6956,
309
+ "step": 500
310
+ },
311
+ {
312
+ "epoch": 0.38,
313
+ "learning_rate": 4.36519790888723e-05,
314
+ "loss": 1.689,
315
+ "step": 510
316
+ },
317
+ {
318
+ "epoch": 0.39,
319
+ "learning_rate": 4.352750809061489e-05,
320
+ "loss": 1.673,
321
+ "step": 520
322
+ },
323
+ {
324
+ "epoch": 0.4,
325
+ "learning_rate": 4.340303709235748e-05,
326
+ "loss": 1.6968,
327
+ "step": 530
328
+ },
329
+ {
330
+ "epoch": 0.4,
331
+ "learning_rate": 4.327856609410008e-05,
332
+ "loss": 1.6569,
333
+ "step": 540
334
+ },
335
+ {
336
+ "epoch": 0.41,
337
+ "learning_rate": 4.315409509584267e-05,
338
+ "loss": 1.6777,
339
+ "step": 550
340
+ },
341
+ {
342
+ "epoch": 0.42,
343
+ "learning_rate": 4.302962409758526e-05,
344
+ "loss": 1.71,
345
+ "step": 560
346
+ },
347
+ {
348
+ "epoch": 0.43,
349
+ "learning_rate": 4.290515309932786e-05,
350
+ "loss": 1.6943,
351
+ "step": 570
352
+ },
353
+ {
354
+ "epoch": 0.43,
355
+ "learning_rate": 4.278068210107045e-05,
356
+ "loss": 1.6925,
357
+ "step": 580
358
+ },
359
+ {
360
+ "epoch": 0.44,
361
+ "learning_rate": 4.2656211102813046e-05,
362
+ "loss": 1.6628,
363
+ "step": 590
364
+ },
365
+ {
366
+ "epoch": 0.45,
367
+ "learning_rate": 4.253174010455564e-05,
368
+ "loss": 1.7123,
369
+ "step": 600
370
+ },
371
+ {
372
+ "epoch": 0.46,
373
+ "learning_rate": 4.2407269106298235e-05,
374
+ "loss": 1.6772,
375
+ "step": 610
376
+ },
377
+ {
378
+ "epoch": 0.46,
379
+ "learning_rate": 4.228279810804083e-05,
380
+ "loss": 1.6429,
381
+ "step": 620
382
+ },
383
+ {
384
+ "epoch": 0.47,
385
+ "learning_rate": 4.2158327109783425e-05,
386
+ "loss": 1.6472,
387
+ "step": 630
388
+ },
389
+ {
390
+ "epoch": 0.48,
391
+ "learning_rate": 4.203385611152601e-05,
392
+ "loss": 1.6825,
393
+ "step": 640
394
+ },
395
+ {
396
+ "epoch": 0.49,
397
+ "learning_rate": 4.1909385113268615e-05,
398
+ "loss": 1.6825,
399
+ "step": 650
400
+ },
401
+ {
402
+ "epoch": 0.49,
403
+ "learning_rate": 4.17849141150112e-05,
404
+ "loss": 1.6818,
405
+ "step": 660
406
+ },
407
+ {
408
+ "epoch": 0.5,
409
+ "learning_rate": 4.16604431167538e-05,
410
+ "loss": 1.7325,
411
+ "step": 670
412
+ },
413
+ {
414
+ "epoch": 0.51,
415
+ "learning_rate": 4.153597211849639e-05,
416
+ "loss": 1.7008,
417
+ "step": 680
418
+ },
419
+ {
420
+ "epoch": 0.52,
421
+ "learning_rate": 4.141150112023899e-05,
422
+ "loss": 1.655,
423
+ "step": 690
424
+ },
425
+ {
426
+ "epoch": 0.52,
427
+ "learning_rate": 4.128703012198158e-05,
428
+ "loss": 1.6718,
429
+ "step": 700
430
+ },
431
+ {
432
+ "epoch": 0.53,
433
+ "learning_rate": 4.116255912372418e-05,
434
+ "loss": 1.6412,
435
+ "step": 710
436
+ },
437
+ {
438
+ "epoch": 0.54,
439
+ "learning_rate": 4.1038088125466765e-05,
440
+ "loss": 1.7091,
441
+ "step": 720
442
+ },
443
+ {
444
+ "epoch": 0.55,
445
+ "learning_rate": 4.091361712720937e-05,
446
+ "loss": 1.726,
447
+ "step": 730
448
+ },
449
+ {
450
+ "epoch": 0.55,
451
+ "learning_rate": 4.0789146128951955e-05,
452
+ "loss": 1.6655,
453
+ "step": 740
454
+ },
455
+ {
456
+ "epoch": 0.56,
457
+ "learning_rate": 4.066467513069455e-05,
458
+ "loss": 1.7242,
459
+ "step": 750
460
+ },
461
+ {
462
+ "epoch": 0.57,
463
+ "learning_rate": 4.0540204132437145e-05,
464
+ "loss": 1.6714,
465
+ "step": 760
466
+ },
467
+ {
468
+ "epoch": 0.58,
469
+ "learning_rate": 4.041573313417974e-05,
470
+ "loss": 1.6646,
471
+ "step": 770
472
+ },
473
+ {
474
+ "epoch": 0.58,
475
+ "learning_rate": 4.029126213592233e-05,
476
+ "loss": 1.6809,
477
+ "step": 780
478
+ },
479
+ {
480
+ "epoch": 0.59,
481
+ "learning_rate": 4.016679113766493e-05,
482
+ "loss": 1.6868,
483
+ "step": 790
484
+ },
485
+ {
486
+ "epoch": 0.6,
487
+ "learning_rate": 4.004232013940752e-05,
488
+ "loss": 1.694,
489
+ "step": 800
490
+ },
491
+ {
492
+ "epoch": 0.6,
493
+ "learning_rate": 3.991784914115011e-05,
494
+ "loss": 1.6801,
495
+ "step": 810
496
+ },
497
+ {
498
+ "epoch": 0.61,
499
+ "learning_rate": 3.979337814289271e-05,
500
+ "loss": 1.6861,
501
+ "step": 820
502
+ },
503
+ {
504
+ "epoch": 0.62,
505
+ "learning_rate": 3.96689071446353e-05,
506
+ "loss": 1.6646,
507
+ "step": 830
508
+ },
509
+ {
510
+ "epoch": 0.63,
511
+ "learning_rate": 3.95444361463779e-05,
512
+ "loss": 1.6764,
513
+ "step": 840
514
+ },
515
+ {
516
+ "epoch": 0.63,
517
+ "learning_rate": 3.941996514812049e-05,
518
+ "loss": 1.6821,
519
+ "step": 850
520
+ },
521
+ {
522
+ "epoch": 0.64,
523
+ "learning_rate": 3.929549414986308e-05,
524
+ "loss": 1.6681,
525
+ "step": 860
526
+ },
527
+ {
528
+ "epoch": 0.65,
529
+ "learning_rate": 3.917102315160568e-05,
530
+ "loss": 1.6847,
531
+ "step": 870
532
+ },
533
+ {
534
+ "epoch": 0.66,
535
+ "learning_rate": 3.904655215334827e-05,
536
+ "loss": 1.6084,
537
+ "step": 880
538
+ },
539
+ {
540
+ "epoch": 0.66,
541
+ "learning_rate": 3.8922081155090865e-05,
542
+ "loss": 1.6624,
543
+ "step": 890
544
+ },
545
+ {
546
+ "epoch": 0.67,
547
+ "learning_rate": 3.879761015683346e-05,
548
+ "loss": 1.6619,
549
+ "step": 900
550
+ },
551
+ {
552
+ "epoch": 0.68,
553
+ "learning_rate": 3.8673139158576054e-05,
554
+ "loss": 1.7024,
555
+ "step": 910
556
+ },
557
+ {
558
+ "epoch": 0.69,
559
+ "learning_rate": 3.854866816031865e-05,
560
+ "loss": 1.7301,
561
+ "step": 920
562
+ },
563
+ {
564
+ "epoch": 0.69,
565
+ "learning_rate": 3.8424197162061244e-05,
566
+ "loss": 1.6907,
567
+ "step": 930
568
+ },
569
+ {
570
+ "epoch": 0.7,
571
+ "learning_rate": 3.829972616380383e-05,
572
+ "loss": 1.7046,
573
+ "step": 940
574
+ },
575
+ {
576
+ "epoch": 0.71,
577
+ "learning_rate": 3.8175255165546434e-05,
578
+ "loss": 1.6661,
579
+ "step": 950
580
+ },
581
+ {
582
+ "epoch": 0.72,
583
+ "learning_rate": 3.805078416728902e-05,
584
+ "loss": 1.6981,
585
+ "step": 960
586
+ },
587
+ {
588
+ "epoch": 0.72,
589
+ "learning_rate": 3.792631316903162e-05,
590
+ "loss": 1.662,
591
+ "step": 970
592
+ },
593
+ {
594
+ "epoch": 0.73,
595
+ "learning_rate": 3.780184217077421e-05,
596
+ "loss": 1.7073,
597
+ "step": 980
598
+ },
599
+ {
600
+ "epoch": 0.74,
601
+ "learning_rate": 3.7677371172516806e-05,
602
+ "loss": 1.6488,
603
+ "step": 990
604
+ },
605
+ {
606
+ "epoch": 0.75,
607
+ "learning_rate": 3.7552900174259395e-05,
608
+ "loss": 1.6708,
609
+ "step": 1000
610
+ }
611
+ ],
612
+ "logging_steps": 10,
613
+ "max_steps": 4017,
614
+ "num_input_tokens_seen": 0,
615
+ "num_train_epochs": 3,
616
+ "save_steps": 500,
617
+ "total_flos": 6.3202482192384e+17,
618
+ "train_batch_size": 4,
619
+ "trial_name": null,
620
+ "trial_params": null
621
+ }
119_Llama-2-13b-hf/checkpoint-1000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:95fdd19e2ed6f33177f737bfa3ccdb7380f84c750beb151600638af981e7df02
3
+ size 4283
119_Llama-2-13b-hf/checkpoint-1500/README.md ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+ ### Framework versions
7
+
8
+
9
+ - PEFT 0.4.0
119_Llama-2-13b-hf/checkpoint-1500/adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "meta-llama/Llama-2-13b-hf",
4
+ "bias": "none",
5
+ "fan_in_fan_out": false,
6
+ "inference_mode": true,
7
+ "init_lora_weights": true,
8
+ "layers_pattern": null,
9
+ "layers_to_transform": null,
10
+ "lora_alpha": 32,
11
+ "lora_dropout": 0.05,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 8,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "q_proj",
18
+ "v_proj"
19
+ ],
20
+ "task_type": "CAUSAL_LM"
21
+ }
119_Llama-2-13b-hf/checkpoint-1500/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bff574f1f26c2035d9ea542d7303bf40b85470db80e825d0479e8692671fc6d9
3
+ size 26235704
119_Llama-2-13b-hf/checkpoint-1500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9c1e2dbbcbc8f6697727d7d5346bd719aa7aca73f52cc14c2048af257e60e5c6
3
+ size 52567237
119_Llama-2-13b-hf/checkpoint-1500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bae61c7ef42fa8f02e89a1d61f60fe6bf777cfc5a1f43edb62e9848d140b70fb
3
+ size 14575
119_Llama-2-13b-hf/checkpoint-1500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:99dd10fd675d3d53d901469e241efe6beb3d8631e1994f23c1c41a8704ab425c
3
+ size 627
119_Llama-2-13b-hf/checkpoint-1500/trainer_state.json ADDED
@@ -0,0 +1,921 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.1202389843166543,
5
+ "eval_steps": 500,
6
+ "global_step": 1500,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.01,
13
+ "learning_rate": 4.98755290017426e-05,
14
+ "loss": 1.7891,
15
+ "step": 10
16
+ },
17
+ {
18
+ "epoch": 0.01,
19
+ "learning_rate": 4.975105800348519e-05,
20
+ "loss": 1.7869,
21
+ "step": 20
22
+ },
23
+ {
24
+ "epoch": 0.02,
25
+ "learning_rate": 4.962658700522779e-05,
26
+ "loss": 1.7868,
27
+ "step": 30
28
+ },
29
+ {
30
+ "epoch": 0.03,
31
+ "learning_rate": 4.9502116006970375e-05,
32
+ "loss": 1.7547,
33
+ "step": 40
34
+ },
35
+ {
36
+ "epoch": 0.04,
37
+ "learning_rate": 4.937764500871298e-05,
38
+ "loss": 1.6772,
39
+ "step": 50
40
+ },
41
+ {
42
+ "epoch": 0.04,
43
+ "learning_rate": 4.9253174010455565e-05,
44
+ "loss": 1.7076,
45
+ "step": 60
46
+ },
47
+ {
48
+ "epoch": 0.05,
49
+ "learning_rate": 4.912870301219816e-05,
50
+ "loss": 1.6948,
51
+ "step": 70
52
+ },
53
+ {
54
+ "epoch": 0.06,
55
+ "learning_rate": 4.9004232013940755e-05,
56
+ "loss": 1.7135,
57
+ "step": 80
58
+ },
59
+ {
60
+ "epoch": 0.07,
61
+ "learning_rate": 4.887976101568335e-05,
62
+ "loss": 1.7007,
63
+ "step": 90
64
+ },
65
+ {
66
+ "epoch": 0.07,
67
+ "learning_rate": 4.8755290017425944e-05,
68
+ "loss": 1.7247,
69
+ "step": 100
70
+ },
71
+ {
72
+ "epoch": 0.08,
73
+ "learning_rate": 4.863081901916854e-05,
74
+ "loss": 1.719,
75
+ "step": 110
76
+ },
77
+ {
78
+ "epoch": 0.09,
79
+ "learning_rate": 4.850634802091113e-05,
80
+ "loss": 1.7209,
81
+ "step": 120
82
+ },
83
+ {
84
+ "epoch": 0.1,
85
+ "learning_rate": 4.838187702265373e-05,
86
+ "loss": 1.7354,
87
+ "step": 130
88
+ },
89
+ {
90
+ "epoch": 0.1,
91
+ "learning_rate": 4.825740602439632e-05,
92
+ "loss": 1.7025,
93
+ "step": 140
94
+ },
95
+ {
96
+ "epoch": 0.11,
97
+ "learning_rate": 4.813293502613891e-05,
98
+ "loss": 1.6664,
99
+ "step": 150
100
+ },
101
+ {
102
+ "epoch": 0.12,
103
+ "learning_rate": 4.800846402788151e-05,
104
+ "loss": 1.6851,
105
+ "step": 160
106
+ },
107
+ {
108
+ "epoch": 0.13,
109
+ "learning_rate": 4.78839930296241e-05,
110
+ "loss": 1.6749,
111
+ "step": 170
112
+ },
113
+ {
114
+ "epoch": 0.13,
115
+ "learning_rate": 4.7759522031366697e-05,
116
+ "loss": 1.6887,
117
+ "step": 180
118
+ },
119
+ {
120
+ "epoch": 0.14,
121
+ "learning_rate": 4.763505103310929e-05,
122
+ "loss": 1.6571,
123
+ "step": 190
124
+ },
125
+ {
126
+ "epoch": 0.15,
127
+ "learning_rate": 4.751058003485188e-05,
128
+ "loss": 1.7136,
129
+ "step": 200
130
+ },
131
+ {
132
+ "epoch": 0.16,
133
+ "learning_rate": 4.738610903659448e-05,
134
+ "loss": 1.7098,
135
+ "step": 210
136
+ },
137
+ {
138
+ "epoch": 0.16,
139
+ "learning_rate": 4.726163803833707e-05,
140
+ "loss": 1.7285,
141
+ "step": 220
142
+ },
143
+ {
144
+ "epoch": 0.17,
145
+ "learning_rate": 4.7137167040079664e-05,
146
+ "loss": 1.7038,
147
+ "step": 230
148
+ },
149
+ {
150
+ "epoch": 0.18,
151
+ "learning_rate": 4.701269604182226e-05,
152
+ "loss": 1.692,
153
+ "step": 240
154
+ },
155
+ {
156
+ "epoch": 0.19,
157
+ "learning_rate": 4.6888225043564854e-05,
158
+ "loss": 1.6825,
159
+ "step": 250
160
+ },
161
+ {
162
+ "epoch": 0.19,
163
+ "learning_rate": 4.676375404530744e-05,
164
+ "loss": 1.7254,
165
+ "step": 260
166
+ },
167
+ {
168
+ "epoch": 0.2,
169
+ "learning_rate": 4.6639283047050044e-05,
170
+ "loss": 1.6765,
171
+ "step": 270
172
+ },
173
+ {
174
+ "epoch": 0.21,
175
+ "learning_rate": 4.651481204879263e-05,
176
+ "loss": 1.6796,
177
+ "step": 280
178
+ },
179
+ {
180
+ "epoch": 0.22,
181
+ "learning_rate": 4.6390341050535227e-05,
182
+ "loss": 1.7302,
183
+ "step": 290
184
+ },
185
+ {
186
+ "epoch": 0.22,
187
+ "learning_rate": 4.626587005227782e-05,
188
+ "loss": 1.7469,
189
+ "step": 300
190
+ },
191
+ {
192
+ "epoch": 0.23,
193
+ "learning_rate": 4.6141399054020416e-05,
194
+ "loss": 1.6345,
195
+ "step": 310
196
+ },
197
+ {
198
+ "epoch": 0.24,
199
+ "learning_rate": 4.601692805576301e-05,
200
+ "loss": 1.6761,
201
+ "step": 320
202
+ },
203
+ {
204
+ "epoch": 0.25,
205
+ "learning_rate": 4.5892457057505606e-05,
206
+ "loss": 1.6602,
207
+ "step": 330
208
+ },
209
+ {
210
+ "epoch": 0.25,
211
+ "learning_rate": 4.5767986059248194e-05,
212
+ "loss": 1.6637,
213
+ "step": 340
214
+ },
215
+ {
216
+ "epoch": 0.26,
217
+ "learning_rate": 4.5643515060990796e-05,
218
+ "loss": 1.6953,
219
+ "step": 350
220
+ },
221
+ {
222
+ "epoch": 0.27,
223
+ "learning_rate": 4.5519044062733384e-05,
224
+ "loss": 1.6829,
225
+ "step": 360
226
+ },
227
+ {
228
+ "epoch": 0.28,
229
+ "learning_rate": 4.539457306447598e-05,
230
+ "loss": 1.657,
231
+ "step": 370
232
+ },
233
+ {
234
+ "epoch": 0.28,
235
+ "learning_rate": 4.5270102066218574e-05,
236
+ "loss": 1.647,
237
+ "step": 380
238
+ },
239
+ {
240
+ "epoch": 0.29,
241
+ "learning_rate": 4.514563106796117e-05,
242
+ "loss": 1.6854,
243
+ "step": 390
244
+ },
245
+ {
246
+ "epoch": 0.3,
247
+ "learning_rate": 4.502116006970376e-05,
248
+ "loss": 1.6737,
249
+ "step": 400
250
+ },
251
+ {
252
+ "epoch": 0.31,
253
+ "learning_rate": 4.489668907144636e-05,
254
+ "loss": 1.6938,
255
+ "step": 410
256
+ },
257
+ {
258
+ "epoch": 0.31,
259
+ "learning_rate": 4.4772218073188946e-05,
260
+ "loss": 1.6721,
261
+ "step": 420
262
+ },
263
+ {
264
+ "epoch": 0.32,
265
+ "learning_rate": 4.464774707493155e-05,
266
+ "loss": 1.6683,
267
+ "step": 430
268
+ },
269
+ {
270
+ "epoch": 0.33,
271
+ "learning_rate": 4.4523276076674136e-05,
272
+ "loss": 1.6229,
273
+ "step": 440
274
+ },
275
+ {
276
+ "epoch": 0.34,
277
+ "learning_rate": 4.439880507841673e-05,
278
+ "loss": 1.6593,
279
+ "step": 450
280
+ },
281
+ {
282
+ "epoch": 0.34,
283
+ "learning_rate": 4.4274334080159326e-05,
284
+ "loss": 1.6844,
285
+ "step": 460
286
+ },
287
+ {
288
+ "epoch": 0.35,
289
+ "learning_rate": 4.414986308190192e-05,
290
+ "loss": 1.6754,
291
+ "step": 470
292
+ },
293
+ {
294
+ "epoch": 0.36,
295
+ "learning_rate": 4.4025392083644516e-05,
296
+ "loss": 1.6747,
297
+ "step": 480
298
+ },
299
+ {
300
+ "epoch": 0.37,
301
+ "learning_rate": 4.390092108538711e-05,
302
+ "loss": 1.6789,
303
+ "step": 490
304
+ },
305
+ {
306
+ "epoch": 0.37,
307
+ "learning_rate": 4.37764500871297e-05,
308
+ "loss": 1.6956,
309
+ "step": 500
310
+ },
311
+ {
312
+ "epoch": 0.38,
313
+ "learning_rate": 4.36519790888723e-05,
314
+ "loss": 1.689,
315
+ "step": 510
316
+ },
317
+ {
318
+ "epoch": 0.39,
319
+ "learning_rate": 4.352750809061489e-05,
320
+ "loss": 1.673,
321
+ "step": 520
322
+ },
323
+ {
324
+ "epoch": 0.4,
325
+ "learning_rate": 4.340303709235748e-05,
326
+ "loss": 1.6968,
327
+ "step": 530
328
+ },
329
+ {
330
+ "epoch": 0.4,
331
+ "learning_rate": 4.327856609410008e-05,
332
+ "loss": 1.6569,
333
+ "step": 540
334
+ },
335
+ {
336
+ "epoch": 0.41,
337
+ "learning_rate": 4.315409509584267e-05,
338
+ "loss": 1.6777,
339
+ "step": 550
340
+ },
341
+ {
342
+ "epoch": 0.42,
343
+ "learning_rate": 4.302962409758526e-05,
344
+ "loss": 1.71,
345
+ "step": 560
346
+ },
347
+ {
348
+ "epoch": 0.43,
349
+ "learning_rate": 4.290515309932786e-05,
350
+ "loss": 1.6943,
351
+ "step": 570
352
+ },
353
+ {
354
+ "epoch": 0.43,
355
+ "learning_rate": 4.278068210107045e-05,
356
+ "loss": 1.6925,
357
+ "step": 580
358
+ },
359
+ {
360
+ "epoch": 0.44,
361
+ "learning_rate": 4.2656211102813046e-05,
362
+ "loss": 1.6628,
363
+ "step": 590
364
+ },
365
+ {
366
+ "epoch": 0.45,
367
+ "learning_rate": 4.253174010455564e-05,
368
+ "loss": 1.7123,
369
+ "step": 600
370
+ },
371
+ {
372
+ "epoch": 0.46,
373
+ "learning_rate": 4.2407269106298235e-05,
374
+ "loss": 1.6772,
375
+ "step": 610
376
+ },
377
+ {
378
+ "epoch": 0.46,
379
+ "learning_rate": 4.228279810804083e-05,
380
+ "loss": 1.6429,
381
+ "step": 620
382
+ },
383
+ {
384
+ "epoch": 0.47,
385
+ "learning_rate": 4.2158327109783425e-05,
386
+ "loss": 1.6472,
387
+ "step": 630
388
+ },
389
+ {
390
+ "epoch": 0.48,
391
+ "learning_rate": 4.203385611152601e-05,
392
+ "loss": 1.6825,
393
+ "step": 640
394
+ },
395
+ {
396
+ "epoch": 0.49,
397
+ "learning_rate": 4.1909385113268615e-05,
398
+ "loss": 1.6825,
399
+ "step": 650
400
+ },
401
+ {
402
+ "epoch": 0.49,
403
+ "learning_rate": 4.17849141150112e-05,
404
+ "loss": 1.6818,
405
+ "step": 660
406
+ },
407
+ {
408
+ "epoch": 0.5,
409
+ "learning_rate": 4.16604431167538e-05,
410
+ "loss": 1.7325,
411
+ "step": 670
412
+ },
413
+ {
414
+ "epoch": 0.51,
415
+ "learning_rate": 4.153597211849639e-05,
416
+ "loss": 1.7008,
417
+ "step": 680
418
+ },
419
+ {
420
+ "epoch": 0.52,
421
+ "learning_rate": 4.141150112023899e-05,
422
+ "loss": 1.655,
423
+ "step": 690
424
+ },
425
+ {
426
+ "epoch": 0.52,
427
+ "learning_rate": 4.128703012198158e-05,
428
+ "loss": 1.6718,
429
+ "step": 700
430
+ },
431
+ {
432
+ "epoch": 0.53,
433
+ "learning_rate": 4.116255912372418e-05,
434
+ "loss": 1.6412,
435
+ "step": 710
436
+ },
437
+ {
438
+ "epoch": 0.54,
439
+ "learning_rate": 4.1038088125466765e-05,
440
+ "loss": 1.7091,
441
+ "step": 720
442
+ },
443
+ {
444
+ "epoch": 0.55,
445
+ "learning_rate": 4.091361712720937e-05,
446
+ "loss": 1.726,
447
+ "step": 730
448
+ },
449
+ {
450
+ "epoch": 0.55,
451
+ "learning_rate": 4.0789146128951955e-05,
452
+ "loss": 1.6655,
453
+ "step": 740
454
+ },
455
+ {
456
+ "epoch": 0.56,
457
+ "learning_rate": 4.066467513069455e-05,
458
+ "loss": 1.7242,
459
+ "step": 750
460
+ },
461
+ {
462
+ "epoch": 0.57,
463
+ "learning_rate": 4.0540204132437145e-05,
464
+ "loss": 1.6714,
465
+ "step": 760
466
+ },
467
+ {
468
+ "epoch": 0.58,
469
+ "learning_rate": 4.041573313417974e-05,
470
+ "loss": 1.6646,
471
+ "step": 770
472
+ },
473
+ {
474
+ "epoch": 0.58,
475
+ "learning_rate": 4.029126213592233e-05,
476
+ "loss": 1.6809,
477
+ "step": 780
478
+ },
479
+ {
480
+ "epoch": 0.59,
481
+ "learning_rate": 4.016679113766493e-05,
482
+ "loss": 1.6868,
483
+ "step": 790
484
+ },
485
+ {
486
+ "epoch": 0.6,
487
+ "learning_rate": 4.004232013940752e-05,
488
+ "loss": 1.694,
489
+ "step": 800
490
+ },
491
+ {
492
+ "epoch": 0.6,
493
+ "learning_rate": 3.991784914115011e-05,
494
+ "loss": 1.6801,
495
+ "step": 810
496
+ },
497
+ {
498
+ "epoch": 0.61,
499
+ "learning_rate": 3.979337814289271e-05,
500
+ "loss": 1.6861,
501
+ "step": 820
502
+ },
503
+ {
504
+ "epoch": 0.62,
505
+ "learning_rate": 3.96689071446353e-05,
506
+ "loss": 1.6646,
507
+ "step": 830
508
+ },
509
+ {
510
+ "epoch": 0.63,
511
+ "learning_rate": 3.95444361463779e-05,
512
+ "loss": 1.6764,
513
+ "step": 840
514
+ },
515
+ {
516
+ "epoch": 0.63,
517
+ "learning_rate": 3.941996514812049e-05,
518
+ "loss": 1.6821,
519
+ "step": 850
520
+ },
521
+ {
522
+ "epoch": 0.64,
523
+ "learning_rate": 3.929549414986308e-05,
524
+ "loss": 1.6681,
525
+ "step": 860
526
+ },
527
+ {
528
+ "epoch": 0.65,
529
+ "learning_rate": 3.917102315160568e-05,
530
+ "loss": 1.6847,
531
+ "step": 870
532
+ },
533
+ {
534
+ "epoch": 0.66,
535
+ "learning_rate": 3.904655215334827e-05,
536
+ "loss": 1.6084,
537
+ "step": 880
538
+ },
539
+ {
540
+ "epoch": 0.66,
541
+ "learning_rate": 3.8922081155090865e-05,
542
+ "loss": 1.6624,
543
+ "step": 890
544
+ },
545
+ {
546
+ "epoch": 0.67,
547
+ "learning_rate": 3.879761015683346e-05,
548
+ "loss": 1.6619,
549
+ "step": 900
550
+ },
551
+ {
552
+ "epoch": 0.68,
553
+ "learning_rate": 3.8673139158576054e-05,
554
+ "loss": 1.7024,
555
+ "step": 910
556
+ },
557
+ {
558
+ "epoch": 0.69,
559
+ "learning_rate": 3.854866816031865e-05,
560
+ "loss": 1.7301,
561
+ "step": 920
562
+ },
563
+ {
564
+ "epoch": 0.69,
565
+ "learning_rate": 3.8424197162061244e-05,
566
+ "loss": 1.6907,
567
+ "step": 930
568
+ },
569
+ {
570
+ "epoch": 0.7,
571
+ "learning_rate": 3.829972616380383e-05,
572
+ "loss": 1.7046,
573
+ "step": 940
574
+ },
575
+ {
576
+ "epoch": 0.71,
577
+ "learning_rate": 3.8175255165546434e-05,
578
+ "loss": 1.6661,
579
+ "step": 950
580
+ },
581
+ {
582
+ "epoch": 0.72,
583
+ "learning_rate": 3.805078416728902e-05,
584
+ "loss": 1.6981,
585
+ "step": 960
586
+ },
587
+ {
588
+ "epoch": 0.72,
589
+ "learning_rate": 3.792631316903162e-05,
590
+ "loss": 1.662,
591
+ "step": 970
592
+ },
593
+ {
594
+ "epoch": 0.73,
595
+ "learning_rate": 3.780184217077421e-05,
596
+ "loss": 1.7073,
597
+ "step": 980
598
+ },
599
+ {
600
+ "epoch": 0.74,
601
+ "learning_rate": 3.7677371172516806e-05,
602
+ "loss": 1.6488,
603
+ "step": 990
604
+ },
605
+ {
606
+ "epoch": 0.75,
607
+ "learning_rate": 3.7552900174259395e-05,
608
+ "loss": 1.6708,
609
+ "step": 1000
610
+ },
611
+ {
612
+ "epoch": 0.75,
613
+ "learning_rate": 3.7428429176001996e-05,
614
+ "loss": 1.7045,
615
+ "step": 1010
616
+ },
617
+ {
618
+ "epoch": 0.76,
619
+ "learning_rate": 3.7303958177744584e-05,
620
+ "loss": 1.6842,
621
+ "step": 1020
622
+ },
623
+ {
624
+ "epoch": 0.77,
625
+ "learning_rate": 3.717948717948718e-05,
626
+ "loss": 1.6682,
627
+ "step": 1030
628
+ },
629
+ {
630
+ "epoch": 0.78,
631
+ "learning_rate": 3.7055016181229774e-05,
632
+ "loss": 1.636,
633
+ "step": 1040
634
+ },
635
+ {
636
+ "epoch": 0.78,
637
+ "learning_rate": 3.693054518297237e-05,
638
+ "loss": 1.6801,
639
+ "step": 1050
640
+ },
641
+ {
642
+ "epoch": 0.79,
643
+ "learning_rate": 3.6806074184714964e-05,
644
+ "loss": 1.6371,
645
+ "step": 1060
646
+ },
647
+ {
648
+ "epoch": 0.8,
649
+ "learning_rate": 3.668160318645756e-05,
650
+ "loss": 1.6755,
651
+ "step": 1070
652
+ },
653
+ {
654
+ "epoch": 0.81,
655
+ "learning_rate": 3.655713218820015e-05,
656
+ "loss": 1.7046,
657
+ "step": 1080
658
+ },
659
+ {
660
+ "epoch": 0.81,
661
+ "learning_rate": 3.643266118994275e-05,
662
+ "loss": 1.707,
663
+ "step": 1090
664
+ },
665
+ {
666
+ "epoch": 0.82,
667
+ "learning_rate": 3.6308190191685337e-05,
668
+ "loss": 1.6568,
669
+ "step": 1100
670
+ },
671
+ {
672
+ "epoch": 0.83,
673
+ "learning_rate": 3.618371919342793e-05,
674
+ "loss": 1.6447,
675
+ "step": 1110
676
+ },
677
+ {
678
+ "epoch": 0.84,
679
+ "learning_rate": 3.6059248195170526e-05,
680
+ "loss": 1.691,
681
+ "step": 1120
682
+ },
683
+ {
684
+ "epoch": 0.84,
685
+ "learning_rate": 3.593477719691312e-05,
686
+ "loss": 1.6747,
687
+ "step": 1130
688
+ },
689
+ {
690
+ "epoch": 0.85,
691
+ "learning_rate": 3.5810306198655716e-05,
692
+ "loss": 1.656,
693
+ "step": 1140
694
+ },
695
+ {
696
+ "epoch": 0.86,
697
+ "learning_rate": 3.568583520039831e-05,
698
+ "loss": 1.6485,
699
+ "step": 1150
700
+ },
701
+ {
702
+ "epoch": 0.87,
703
+ "learning_rate": 3.55613642021409e-05,
704
+ "loss": 1.642,
705
+ "step": 1160
706
+ },
707
+ {
708
+ "epoch": 0.87,
709
+ "learning_rate": 3.54368932038835e-05,
710
+ "loss": 1.6891,
711
+ "step": 1170
712
+ },
713
+ {
714
+ "epoch": 0.88,
715
+ "learning_rate": 3.531242220562609e-05,
716
+ "loss": 1.716,
717
+ "step": 1180
718
+ },
719
+ {
720
+ "epoch": 0.89,
721
+ "learning_rate": 3.5187951207368684e-05,
722
+ "loss": 1.6785,
723
+ "step": 1190
724
+ },
725
+ {
726
+ "epoch": 0.9,
727
+ "learning_rate": 3.506348020911128e-05,
728
+ "loss": 1.7053,
729
+ "step": 1200
730
+ },
731
+ {
732
+ "epoch": 0.9,
733
+ "learning_rate": 3.493900921085387e-05,
734
+ "loss": 1.6833,
735
+ "step": 1210
736
+ },
737
+ {
738
+ "epoch": 0.91,
739
+ "learning_rate": 3.481453821259646e-05,
740
+ "loss": 1.7336,
741
+ "step": 1220
742
+ },
743
+ {
744
+ "epoch": 0.92,
745
+ "learning_rate": 3.469006721433906e-05,
746
+ "loss": 1.6496,
747
+ "step": 1230
748
+ },
749
+ {
750
+ "epoch": 0.93,
751
+ "learning_rate": 3.456559621608165e-05,
752
+ "loss": 1.7014,
753
+ "step": 1240
754
+ },
755
+ {
756
+ "epoch": 0.93,
757
+ "learning_rate": 3.4441125217824246e-05,
758
+ "loss": 1.6502,
759
+ "step": 1250
760
+ },
761
+ {
762
+ "epoch": 0.94,
763
+ "learning_rate": 3.431665421956684e-05,
764
+ "loss": 1.6617,
765
+ "step": 1260
766
+ },
767
+ {
768
+ "epoch": 0.95,
769
+ "learning_rate": 3.4192183221309436e-05,
770
+ "loss": 1.7155,
771
+ "step": 1270
772
+ },
773
+ {
774
+ "epoch": 0.96,
775
+ "learning_rate": 3.406771222305203e-05,
776
+ "loss": 1.691,
777
+ "step": 1280
778
+ },
779
+ {
780
+ "epoch": 0.96,
781
+ "learning_rate": 3.3943241224794625e-05,
782
+ "loss": 1.6601,
783
+ "step": 1290
784
+ },
785
+ {
786
+ "epoch": 0.97,
787
+ "learning_rate": 3.3818770226537214e-05,
788
+ "loss": 1.6607,
789
+ "step": 1300
790
+ },
791
+ {
792
+ "epoch": 0.98,
793
+ "learning_rate": 3.3694299228279815e-05,
794
+ "loss": 1.7053,
795
+ "step": 1310
796
+ },
797
+ {
798
+ "epoch": 0.99,
799
+ "learning_rate": 3.35698282300224e-05,
800
+ "loss": 1.6681,
801
+ "step": 1320
802
+ },
803
+ {
804
+ "epoch": 0.99,
805
+ "learning_rate": 3.3445357231765e-05,
806
+ "loss": 1.6712,
807
+ "step": 1330
808
+ },
809
+ {
810
+ "epoch": 1.0,
811
+ "learning_rate": 3.332088623350759e-05,
812
+ "loss": 1.6601,
813
+ "step": 1340
814
+ },
815
+ {
816
+ "epoch": 1.01,
817
+ "learning_rate": 3.319641523525019e-05,
818
+ "loss": 1.6522,
819
+ "step": 1350
820
+ },
821
+ {
822
+ "epoch": 1.02,
823
+ "learning_rate": 3.307194423699278e-05,
824
+ "loss": 1.6621,
825
+ "step": 1360
826
+ },
827
+ {
828
+ "epoch": 1.02,
829
+ "learning_rate": 3.294747323873538e-05,
830
+ "loss": 1.6651,
831
+ "step": 1370
832
+ },
833
+ {
834
+ "epoch": 1.03,
835
+ "learning_rate": 3.2823002240477966e-05,
836
+ "loss": 1.6665,
837
+ "step": 1380
838
+ },
839
+ {
840
+ "epoch": 1.04,
841
+ "learning_rate": 3.269853124222057e-05,
842
+ "loss": 1.6811,
843
+ "step": 1390
844
+ },
845
+ {
846
+ "epoch": 1.05,
847
+ "learning_rate": 3.2574060243963156e-05,
848
+ "loss": 1.6632,
849
+ "step": 1400
850
+ },
851
+ {
852
+ "epoch": 1.05,
853
+ "learning_rate": 3.244958924570575e-05,
854
+ "loss": 1.6529,
855
+ "step": 1410
856
+ },
857
+ {
858
+ "epoch": 1.06,
859
+ "learning_rate": 3.2325118247448345e-05,
860
+ "loss": 1.6943,
861
+ "step": 1420
862
+ },
863
+ {
864
+ "epoch": 1.07,
865
+ "learning_rate": 3.220064724919094e-05,
866
+ "loss": 1.6932,
867
+ "step": 1430
868
+ },
869
+ {
870
+ "epoch": 1.08,
871
+ "learning_rate": 3.2076176250933535e-05,
872
+ "loss": 1.6711,
873
+ "step": 1440
874
+ },
875
+ {
876
+ "epoch": 1.08,
877
+ "learning_rate": 3.195170525267613e-05,
878
+ "loss": 1.6711,
879
+ "step": 1450
880
+ },
881
+ {
882
+ "epoch": 1.09,
883
+ "learning_rate": 3.182723425441872e-05,
884
+ "loss": 1.6678,
885
+ "step": 1460
886
+ },
887
+ {
888
+ "epoch": 1.1,
889
+ "learning_rate": 3.170276325616132e-05,
890
+ "loss": 1.7058,
891
+ "step": 1470
892
+ },
893
+ {
894
+ "epoch": 1.11,
895
+ "learning_rate": 3.157829225790391e-05,
896
+ "loss": 1.7017,
897
+ "step": 1480
898
+ },
899
+ {
900
+ "epoch": 1.11,
901
+ "learning_rate": 3.14538212596465e-05,
902
+ "loss": 1.6862,
903
+ "step": 1490
904
+ },
905
+ {
906
+ "epoch": 1.12,
907
+ "learning_rate": 3.13293502613891e-05,
908
+ "loss": 1.6425,
909
+ "step": 1500
910
+ }
911
+ ],
912
+ "logging_steps": 10,
913
+ "max_steps": 4017,
914
+ "num_input_tokens_seen": 0,
915
+ "num_train_epochs": 3,
916
+ "save_steps": 500,
917
+ "total_flos": 9.4803723288576e+17,
918
+ "train_batch_size": 4,
919
+ "trial_name": null,
920
+ "trial_params": null
921
+ }
119_Llama-2-13b-hf/checkpoint-1500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:95fdd19e2ed6f33177f737bfa3ccdb7380f84c750beb151600638af981e7df02
3
+ size 4283
119_Llama-2-13b-hf/checkpoint-2000/README.md ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+ ### Framework versions
7
+
8
+
9
+ - PEFT 0.4.0
119_Llama-2-13b-hf/checkpoint-2000/adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "meta-llama/Llama-2-13b-hf",
4
+ "bias": "none",
5
+ "fan_in_fan_out": false,
6
+ "inference_mode": true,
7
+ "init_lora_weights": true,
8
+ "layers_pattern": null,
9
+ "layers_to_transform": null,
10
+ "lora_alpha": 32,
11
+ "lora_dropout": 0.05,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 8,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "q_proj",
18
+ "v_proj"
19
+ ],
20
+ "task_type": "CAUSAL_LM"
21
+ }
119_Llama-2-13b-hf/checkpoint-2000/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7b430d9cdba58662e3d80bb0853b98de7119ba81195188f0a1271db49cb02d91
3
+ size 26235704
119_Llama-2-13b-hf/checkpoint-2000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:49e58696a5db71c6a50d438b6aa8506b2d68baf023bd2e62edb383ef64967f76
3
+ size 52567237
119_Llama-2-13b-hf/checkpoint-2000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:edb27d360f67aed465d2ad5b657c4c1f1ea2352da31d5c56c984aa8b231c772a
3
+ size 14575
119_Llama-2-13b-hf/checkpoint-2000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f7663f94f338880714c8bb1d60e56090e55323961ef38ee760824cd29bf00518
3
+ size 627
119_Llama-2-13b-hf/checkpoint-2000/trainer_state.json ADDED
@@ -0,0 +1,1221 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.4936519790888723,
5
+ "eval_steps": 500,
6
+ "global_step": 2000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.01,
13
+ "learning_rate": 4.98755290017426e-05,
14
+ "loss": 1.7891,
15
+ "step": 10
16
+ },
17
+ {
18
+ "epoch": 0.01,
19
+ "learning_rate": 4.975105800348519e-05,
20
+ "loss": 1.7869,
21
+ "step": 20
22
+ },
23
+ {
24
+ "epoch": 0.02,
25
+ "learning_rate": 4.962658700522779e-05,
26
+ "loss": 1.7868,
27
+ "step": 30
28
+ },
29
+ {
30
+ "epoch": 0.03,
31
+ "learning_rate": 4.9502116006970375e-05,
32
+ "loss": 1.7547,
33
+ "step": 40
34
+ },
35
+ {
36
+ "epoch": 0.04,
37
+ "learning_rate": 4.937764500871298e-05,
38
+ "loss": 1.6772,
39
+ "step": 50
40
+ },
41
+ {
42
+ "epoch": 0.04,
43
+ "learning_rate": 4.9253174010455565e-05,
44
+ "loss": 1.7076,
45
+ "step": 60
46
+ },
47
+ {
48
+ "epoch": 0.05,
49
+ "learning_rate": 4.912870301219816e-05,
50
+ "loss": 1.6948,
51
+ "step": 70
52
+ },
53
+ {
54
+ "epoch": 0.06,
55
+ "learning_rate": 4.9004232013940755e-05,
56
+ "loss": 1.7135,
57
+ "step": 80
58
+ },
59
+ {
60
+ "epoch": 0.07,
61
+ "learning_rate": 4.887976101568335e-05,
62
+ "loss": 1.7007,
63
+ "step": 90
64
+ },
65
+ {
66
+ "epoch": 0.07,
67
+ "learning_rate": 4.8755290017425944e-05,
68
+ "loss": 1.7247,
69
+ "step": 100
70
+ },
71
+ {
72
+ "epoch": 0.08,
73
+ "learning_rate": 4.863081901916854e-05,
74
+ "loss": 1.719,
75
+ "step": 110
76
+ },
77
+ {
78
+ "epoch": 0.09,
79
+ "learning_rate": 4.850634802091113e-05,
80
+ "loss": 1.7209,
81
+ "step": 120
82
+ },
83
+ {
84
+ "epoch": 0.1,
85
+ "learning_rate": 4.838187702265373e-05,
86
+ "loss": 1.7354,
87
+ "step": 130
88
+ },
89
+ {
90
+ "epoch": 0.1,
91
+ "learning_rate": 4.825740602439632e-05,
92
+ "loss": 1.7025,
93
+ "step": 140
94
+ },
95
+ {
96
+ "epoch": 0.11,
97
+ "learning_rate": 4.813293502613891e-05,
98
+ "loss": 1.6664,
99
+ "step": 150
100
+ },
101
+ {
102
+ "epoch": 0.12,
103
+ "learning_rate": 4.800846402788151e-05,
104
+ "loss": 1.6851,
105
+ "step": 160
106
+ },
107
+ {
108
+ "epoch": 0.13,
109
+ "learning_rate": 4.78839930296241e-05,
110
+ "loss": 1.6749,
111
+ "step": 170
112
+ },
113
+ {
114
+ "epoch": 0.13,
115
+ "learning_rate": 4.7759522031366697e-05,
116
+ "loss": 1.6887,
117
+ "step": 180
118
+ },
119
+ {
120
+ "epoch": 0.14,
121
+ "learning_rate": 4.763505103310929e-05,
122
+ "loss": 1.6571,
123
+ "step": 190
124
+ },
125
+ {
126
+ "epoch": 0.15,
127
+ "learning_rate": 4.751058003485188e-05,
128
+ "loss": 1.7136,
129
+ "step": 200
130
+ },
131
+ {
132
+ "epoch": 0.16,
133
+ "learning_rate": 4.738610903659448e-05,
134
+ "loss": 1.7098,
135
+ "step": 210
136
+ },
137
+ {
138
+ "epoch": 0.16,
139
+ "learning_rate": 4.726163803833707e-05,
140
+ "loss": 1.7285,
141
+ "step": 220
142
+ },
143
+ {
144
+ "epoch": 0.17,
145
+ "learning_rate": 4.7137167040079664e-05,
146
+ "loss": 1.7038,
147
+ "step": 230
148
+ },
149
+ {
150
+ "epoch": 0.18,
151
+ "learning_rate": 4.701269604182226e-05,
152
+ "loss": 1.692,
153
+ "step": 240
154
+ },
155
+ {
156
+ "epoch": 0.19,
157
+ "learning_rate": 4.6888225043564854e-05,
158
+ "loss": 1.6825,
159
+ "step": 250
160
+ },
161
+ {
162
+ "epoch": 0.19,
163
+ "learning_rate": 4.676375404530744e-05,
164
+ "loss": 1.7254,
165
+ "step": 260
166
+ },
167
+ {
168
+ "epoch": 0.2,
169
+ "learning_rate": 4.6639283047050044e-05,
170
+ "loss": 1.6765,
171
+ "step": 270
172
+ },
173
+ {
174
+ "epoch": 0.21,
175
+ "learning_rate": 4.651481204879263e-05,
176
+ "loss": 1.6796,
177
+ "step": 280
178
+ },
179
+ {
180
+ "epoch": 0.22,
181
+ "learning_rate": 4.6390341050535227e-05,
182
+ "loss": 1.7302,
183
+ "step": 290
184
+ },
185
+ {
186
+ "epoch": 0.22,
187
+ "learning_rate": 4.626587005227782e-05,
188
+ "loss": 1.7469,
189
+ "step": 300
190
+ },
191
+ {
192
+ "epoch": 0.23,
193
+ "learning_rate": 4.6141399054020416e-05,
194
+ "loss": 1.6345,
195
+ "step": 310
196
+ },
197
+ {
198
+ "epoch": 0.24,
199
+ "learning_rate": 4.601692805576301e-05,
200
+ "loss": 1.6761,
201
+ "step": 320
202
+ },
203
+ {
204
+ "epoch": 0.25,
205
+ "learning_rate": 4.5892457057505606e-05,
206
+ "loss": 1.6602,
207
+ "step": 330
208
+ },
209
+ {
210
+ "epoch": 0.25,
211
+ "learning_rate": 4.5767986059248194e-05,
212
+ "loss": 1.6637,
213
+ "step": 340
214
+ },
215
+ {
216
+ "epoch": 0.26,
217
+ "learning_rate": 4.5643515060990796e-05,
218
+ "loss": 1.6953,
219
+ "step": 350
220
+ },
221
+ {
222
+ "epoch": 0.27,
223
+ "learning_rate": 4.5519044062733384e-05,
224
+ "loss": 1.6829,
225
+ "step": 360
226
+ },
227
+ {
228
+ "epoch": 0.28,
229
+ "learning_rate": 4.539457306447598e-05,
230
+ "loss": 1.657,
231
+ "step": 370
232
+ },
233
+ {
234
+ "epoch": 0.28,
235
+ "learning_rate": 4.5270102066218574e-05,
236
+ "loss": 1.647,
237
+ "step": 380
238
+ },
239
+ {
240
+ "epoch": 0.29,
241
+ "learning_rate": 4.514563106796117e-05,
242
+ "loss": 1.6854,
243
+ "step": 390
244
+ },
245
+ {
246
+ "epoch": 0.3,
247
+ "learning_rate": 4.502116006970376e-05,
248
+ "loss": 1.6737,
249
+ "step": 400
250
+ },
251
+ {
252
+ "epoch": 0.31,
253
+ "learning_rate": 4.489668907144636e-05,
254
+ "loss": 1.6938,
255
+ "step": 410
256
+ },
257
+ {
258
+ "epoch": 0.31,
259
+ "learning_rate": 4.4772218073188946e-05,
260
+ "loss": 1.6721,
261
+ "step": 420
262
+ },
263
+ {
264
+ "epoch": 0.32,
265
+ "learning_rate": 4.464774707493155e-05,
266
+ "loss": 1.6683,
267
+ "step": 430
268
+ },
269
+ {
270
+ "epoch": 0.33,
271
+ "learning_rate": 4.4523276076674136e-05,
272
+ "loss": 1.6229,
273
+ "step": 440
274
+ },
275
+ {
276
+ "epoch": 0.34,
277
+ "learning_rate": 4.439880507841673e-05,
278
+ "loss": 1.6593,
279
+ "step": 450
280
+ },
281
+ {
282
+ "epoch": 0.34,
283
+ "learning_rate": 4.4274334080159326e-05,
284
+ "loss": 1.6844,
285
+ "step": 460
286
+ },
287
+ {
288
+ "epoch": 0.35,
289
+ "learning_rate": 4.414986308190192e-05,
290
+ "loss": 1.6754,
291
+ "step": 470
292
+ },
293
+ {
294
+ "epoch": 0.36,
295
+ "learning_rate": 4.4025392083644516e-05,
296
+ "loss": 1.6747,
297
+ "step": 480
298
+ },
299
+ {
300
+ "epoch": 0.37,
301
+ "learning_rate": 4.390092108538711e-05,
302
+ "loss": 1.6789,
303
+ "step": 490
304
+ },
305
+ {
306
+ "epoch": 0.37,
307
+ "learning_rate": 4.37764500871297e-05,
308
+ "loss": 1.6956,
309
+ "step": 500
310
+ },
311
+ {
312
+ "epoch": 0.38,
313
+ "learning_rate": 4.36519790888723e-05,
314
+ "loss": 1.689,
315
+ "step": 510
316
+ },
317
+ {
318
+ "epoch": 0.39,
319
+ "learning_rate": 4.352750809061489e-05,
320
+ "loss": 1.673,
321
+ "step": 520
322
+ },
323
+ {
324
+ "epoch": 0.4,
325
+ "learning_rate": 4.340303709235748e-05,
326
+ "loss": 1.6968,
327
+ "step": 530
328
+ },
329
+ {
330
+ "epoch": 0.4,
331
+ "learning_rate": 4.327856609410008e-05,
332
+ "loss": 1.6569,
333
+ "step": 540
334
+ },
335
+ {
336
+ "epoch": 0.41,
337
+ "learning_rate": 4.315409509584267e-05,
338
+ "loss": 1.6777,
339
+ "step": 550
340
+ },
341
+ {
342
+ "epoch": 0.42,
343
+ "learning_rate": 4.302962409758526e-05,
344
+ "loss": 1.71,
345
+ "step": 560
346
+ },
347
+ {
348
+ "epoch": 0.43,
349
+ "learning_rate": 4.290515309932786e-05,
350
+ "loss": 1.6943,
351
+ "step": 570
352
+ },
353
+ {
354
+ "epoch": 0.43,
355
+ "learning_rate": 4.278068210107045e-05,
356
+ "loss": 1.6925,
357
+ "step": 580
358
+ },
359
+ {
360
+ "epoch": 0.44,
361
+ "learning_rate": 4.2656211102813046e-05,
362
+ "loss": 1.6628,
363
+ "step": 590
364
+ },
365
+ {
366
+ "epoch": 0.45,
367
+ "learning_rate": 4.253174010455564e-05,
368
+ "loss": 1.7123,
369
+ "step": 600
370
+ },
371
+ {
372
+ "epoch": 0.46,
373
+ "learning_rate": 4.2407269106298235e-05,
374
+ "loss": 1.6772,
375
+ "step": 610
376
+ },
377
+ {
378
+ "epoch": 0.46,
379
+ "learning_rate": 4.228279810804083e-05,
380
+ "loss": 1.6429,
381
+ "step": 620
382
+ },
383
+ {
384
+ "epoch": 0.47,
385
+ "learning_rate": 4.2158327109783425e-05,
386
+ "loss": 1.6472,
387
+ "step": 630
388
+ },
389
+ {
390
+ "epoch": 0.48,
391
+ "learning_rate": 4.203385611152601e-05,
392
+ "loss": 1.6825,
393
+ "step": 640
394
+ },
395
+ {
396
+ "epoch": 0.49,
397
+ "learning_rate": 4.1909385113268615e-05,
398
+ "loss": 1.6825,
399
+ "step": 650
400
+ },
401
+ {
402
+ "epoch": 0.49,
403
+ "learning_rate": 4.17849141150112e-05,
404
+ "loss": 1.6818,
405
+ "step": 660
406
+ },
407
+ {
408
+ "epoch": 0.5,
409
+ "learning_rate": 4.16604431167538e-05,
410
+ "loss": 1.7325,
411
+ "step": 670
412
+ },
413
+ {
414
+ "epoch": 0.51,
415
+ "learning_rate": 4.153597211849639e-05,
416
+ "loss": 1.7008,
417
+ "step": 680
418
+ },
419
+ {
420
+ "epoch": 0.52,
421
+ "learning_rate": 4.141150112023899e-05,
422
+ "loss": 1.655,
423
+ "step": 690
424
+ },
425
+ {
426
+ "epoch": 0.52,
427
+ "learning_rate": 4.128703012198158e-05,
428
+ "loss": 1.6718,
429
+ "step": 700
430
+ },
431
+ {
432
+ "epoch": 0.53,
433
+ "learning_rate": 4.116255912372418e-05,
434
+ "loss": 1.6412,
435
+ "step": 710
436
+ },
437
+ {
438
+ "epoch": 0.54,
439
+ "learning_rate": 4.1038088125466765e-05,
440
+ "loss": 1.7091,
441
+ "step": 720
442
+ },
443
+ {
444
+ "epoch": 0.55,
445
+ "learning_rate": 4.091361712720937e-05,
446
+ "loss": 1.726,
447
+ "step": 730
448
+ },
449
+ {
450
+ "epoch": 0.55,
451
+ "learning_rate": 4.0789146128951955e-05,
452
+ "loss": 1.6655,
453
+ "step": 740
454
+ },
455
+ {
456
+ "epoch": 0.56,
457
+ "learning_rate": 4.066467513069455e-05,
458
+ "loss": 1.7242,
459
+ "step": 750
460
+ },
461
+ {
462
+ "epoch": 0.57,
463
+ "learning_rate": 4.0540204132437145e-05,
464
+ "loss": 1.6714,
465
+ "step": 760
466
+ },
467
+ {
468
+ "epoch": 0.58,
469
+ "learning_rate": 4.041573313417974e-05,
470
+ "loss": 1.6646,
471
+ "step": 770
472
+ },
473
+ {
474
+ "epoch": 0.58,
475
+ "learning_rate": 4.029126213592233e-05,
476
+ "loss": 1.6809,
477
+ "step": 780
478
+ },
479
+ {
480
+ "epoch": 0.59,
481
+ "learning_rate": 4.016679113766493e-05,
482
+ "loss": 1.6868,
483
+ "step": 790
484
+ },
485
+ {
486
+ "epoch": 0.6,
487
+ "learning_rate": 4.004232013940752e-05,
488
+ "loss": 1.694,
489
+ "step": 800
490
+ },
491
+ {
492
+ "epoch": 0.6,
493
+ "learning_rate": 3.991784914115011e-05,
494
+ "loss": 1.6801,
495
+ "step": 810
496
+ },
497
+ {
498
+ "epoch": 0.61,
499
+ "learning_rate": 3.979337814289271e-05,
500
+ "loss": 1.6861,
501
+ "step": 820
502
+ },
503
+ {
504
+ "epoch": 0.62,
505
+ "learning_rate": 3.96689071446353e-05,
506
+ "loss": 1.6646,
507
+ "step": 830
508
+ },
509
+ {
510
+ "epoch": 0.63,
511
+ "learning_rate": 3.95444361463779e-05,
512
+ "loss": 1.6764,
513
+ "step": 840
514
+ },
515
+ {
516
+ "epoch": 0.63,
517
+ "learning_rate": 3.941996514812049e-05,
518
+ "loss": 1.6821,
519
+ "step": 850
520
+ },
521
+ {
522
+ "epoch": 0.64,
523
+ "learning_rate": 3.929549414986308e-05,
524
+ "loss": 1.6681,
525
+ "step": 860
526
+ },
527
+ {
528
+ "epoch": 0.65,
529
+ "learning_rate": 3.917102315160568e-05,
530
+ "loss": 1.6847,
531
+ "step": 870
532
+ },
533
+ {
534
+ "epoch": 0.66,
535
+ "learning_rate": 3.904655215334827e-05,
536
+ "loss": 1.6084,
537
+ "step": 880
538
+ },
539
+ {
540
+ "epoch": 0.66,
541
+ "learning_rate": 3.8922081155090865e-05,
542
+ "loss": 1.6624,
543
+ "step": 890
544
+ },
545
+ {
546
+ "epoch": 0.67,
547
+ "learning_rate": 3.879761015683346e-05,
548
+ "loss": 1.6619,
549
+ "step": 900
550
+ },
551
+ {
552
+ "epoch": 0.68,
553
+ "learning_rate": 3.8673139158576054e-05,
554
+ "loss": 1.7024,
555
+ "step": 910
556
+ },
557
+ {
558
+ "epoch": 0.69,
559
+ "learning_rate": 3.854866816031865e-05,
560
+ "loss": 1.7301,
561
+ "step": 920
562
+ },
563
+ {
564
+ "epoch": 0.69,
565
+ "learning_rate": 3.8424197162061244e-05,
566
+ "loss": 1.6907,
567
+ "step": 930
568
+ },
569
+ {
570
+ "epoch": 0.7,
571
+ "learning_rate": 3.829972616380383e-05,
572
+ "loss": 1.7046,
573
+ "step": 940
574
+ },
575
+ {
576
+ "epoch": 0.71,
577
+ "learning_rate": 3.8175255165546434e-05,
578
+ "loss": 1.6661,
579
+ "step": 950
580
+ },
581
+ {
582
+ "epoch": 0.72,
583
+ "learning_rate": 3.805078416728902e-05,
584
+ "loss": 1.6981,
585
+ "step": 960
586
+ },
587
+ {
588
+ "epoch": 0.72,
589
+ "learning_rate": 3.792631316903162e-05,
590
+ "loss": 1.662,
591
+ "step": 970
592
+ },
593
+ {
594
+ "epoch": 0.73,
595
+ "learning_rate": 3.780184217077421e-05,
596
+ "loss": 1.7073,
597
+ "step": 980
598
+ },
599
+ {
600
+ "epoch": 0.74,
601
+ "learning_rate": 3.7677371172516806e-05,
602
+ "loss": 1.6488,
603
+ "step": 990
604
+ },
605
+ {
606
+ "epoch": 0.75,
607
+ "learning_rate": 3.7552900174259395e-05,
608
+ "loss": 1.6708,
609
+ "step": 1000
610
+ },
611
+ {
612
+ "epoch": 0.75,
613
+ "learning_rate": 3.7428429176001996e-05,
614
+ "loss": 1.7045,
615
+ "step": 1010
616
+ },
617
+ {
618
+ "epoch": 0.76,
619
+ "learning_rate": 3.7303958177744584e-05,
620
+ "loss": 1.6842,
621
+ "step": 1020
622
+ },
623
+ {
624
+ "epoch": 0.77,
625
+ "learning_rate": 3.717948717948718e-05,
626
+ "loss": 1.6682,
627
+ "step": 1030
628
+ },
629
+ {
630
+ "epoch": 0.78,
631
+ "learning_rate": 3.7055016181229774e-05,
632
+ "loss": 1.636,
633
+ "step": 1040
634
+ },
635
+ {
636
+ "epoch": 0.78,
637
+ "learning_rate": 3.693054518297237e-05,
638
+ "loss": 1.6801,
639
+ "step": 1050
640
+ },
641
+ {
642
+ "epoch": 0.79,
643
+ "learning_rate": 3.6806074184714964e-05,
644
+ "loss": 1.6371,
645
+ "step": 1060
646
+ },
647
+ {
648
+ "epoch": 0.8,
649
+ "learning_rate": 3.668160318645756e-05,
650
+ "loss": 1.6755,
651
+ "step": 1070
652
+ },
653
+ {
654
+ "epoch": 0.81,
655
+ "learning_rate": 3.655713218820015e-05,
656
+ "loss": 1.7046,
657
+ "step": 1080
658
+ },
659
+ {
660
+ "epoch": 0.81,
661
+ "learning_rate": 3.643266118994275e-05,
662
+ "loss": 1.707,
663
+ "step": 1090
664
+ },
665
+ {
666
+ "epoch": 0.82,
667
+ "learning_rate": 3.6308190191685337e-05,
668
+ "loss": 1.6568,
669
+ "step": 1100
670
+ },
671
+ {
672
+ "epoch": 0.83,
673
+ "learning_rate": 3.618371919342793e-05,
674
+ "loss": 1.6447,
675
+ "step": 1110
676
+ },
677
+ {
678
+ "epoch": 0.84,
679
+ "learning_rate": 3.6059248195170526e-05,
680
+ "loss": 1.691,
681
+ "step": 1120
682
+ },
683
+ {
684
+ "epoch": 0.84,
685
+ "learning_rate": 3.593477719691312e-05,
686
+ "loss": 1.6747,
687
+ "step": 1130
688
+ },
689
+ {
690
+ "epoch": 0.85,
691
+ "learning_rate": 3.5810306198655716e-05,
692
+ "loss": 1.656,
693
+ "step": 1140
694
+ },
695
+ {
696
+ "epoch": 0.86,
697
+ "learning_rate": 3.568583520039831e-05,
698
+ "loss": 1.6485,
699
+ "step": 1150
700
+ },
701
+ {
702
+ "epoch": 0.87,
703
+ "learning_rate": 3.55613642021409e-05,
704
+ "loss": 1.642,
705
+ "step": 1160
706
+ },
707
+ {
708
+ "epoch": 0.87,
709
+ "learning_rate": 3.54368932038835e-05,
710
+ "loss": 1.6891,
711
+ "step": 1170
712
+ },
713
+ {
714
+ "epoch": 0.88,
715
+ "learning_rate": 3.531242220562609e-05,
716
+ "loss": 1.716,
717
+ "step": 1180
718
+ },
719
+ {
720
+ "epoch": 0.89,
721
+ "learning_rate": 3.5187951207368684e-05,
722
+ "loss": 1.6785,
723
+ "step": 1190
724
+ },
725
+ {
726
+ "epoch": 0.9,
727
+ "learning_rate": 3.506348020911128e-05,
728
+ "loss": 1.7053,
729
+ "step": 1200
730
+ },
731
+ {
732
+ "epoch": 0.9,
733
+ "learning_rate": 3.493900921085387e-05,
734
+ "loss": 1.6833,
735
+ "step": 1210
736
+ },
737
+ {
738
+ "epoch": 0.91,
739
+ "learning_rate": 3.481453821259646e-05,
740
+ "loss": 1.7336,
741
+ "step": 1220
742
+ },
743
+ {
744
+ "epoch": 0.92,
745
+ "learning_rate": 3.469006721433906e-05,
746
+ "loss": 1.6496,
747
+ "step": 1230
748
+ },
749
+ {
750
+ "epoch": 0.93,
751
+ "learning_rate": 3.456559621608165e-05,
752
+ "loss": 1.7014,
753
+ "step": 1240
754
+ },
755
+ {
756
+ "epoch": 0.93,
757
+ "learning_rate": 3.4441125217824246e-05,
758
+ "loss": 1.6502,
759
+ "step": 1250
760
+ },
761
+ {
762
+ "epoch": 0.94,
763
+ "learning_rate": 3.431665421956684e-05,
764
+ "loss": 1.6617,
765
+ "step": 1260
766
+ },
767
+ {
768
+ "epoch": 0.95,
769
+ "learning_rate": 3.4192183221309436e-05,
770
+ "loss": 1.7155,
771
+ "step": 1270
772
+ },
773
+ {
774
+ "epoch": 0.96,
775
+ "learning_rate": 3.406771222305203e-05,
776
+ "loss": 1.691,
777
+ "step": 1280
778
+ },
779
+ {
780
+ "epoch": 0.96,
781
+ "learning_rate": 3.3943241224794625e-05,
782
+ "loss": 1.6601,
783
+ "step": 1290
784
+ },
785
+ {
786
+ "epoch": 0.97,
787
+ "learning_rate": 3.3818770226537214e-05,
788
+ "loss": 1.6607,
789
+ "step": 1300
790
+ },
791
+ {
792
+ "epoch": 0.98,
793
+ "learning_rate": 3.3694299228279815e-05,
794
+ "loss": 1.7053,
795
+ "step": 1310
796
+ },
797
+ {
798
+ "epoch": 0.99,
799
+ "learning_rate": 3.35698282300224e-05,
800
+ "loss": 1.6681,
801
+ "step": 1320
802
+ },
803
+ {
804
+ "epoch": 0.99,
805
+ "learning_rate": 3.3445357231765e-05,
806
+ "loss": 1.6712,
807
+ "step": 1330
808
+ },
809
+ {
810
+ "epoch": 1.0,
811
+ "learning_rate": 3.332088623350759e-05,
812
+ "loss": 1.6601,
813
+ "step": 1340
814
+ },
815
+ {
816
+ "epoch": 1.01,
817
+ "learning_rate": 3.319641523525019e-05,
818
+ "loss": 1.6522,
819
+ "step": 1350
820
+ },
821
+ {
822
+ "epoch": 1.02,
823
+ "learning_rate": 3.307194423699278e-05,
824
+ "loss": 1.6621,
825
+ "step": 1360
826
+ },
827
+ {
828
+ "epoch": 1.02,
829
+ "learning_rate": 3.294747323873538e-05,
830
+ "loss": 1.6651,
831
+ "step": 1370
832
+ },
833
+ {
834
+ "epoch": 1.03,
835
+ "learning_rate": 3.2823002240477966e-05,
836
+ "loss": 1.6665,
837
+ "step": 1380
838
+ },
839
+ {
840
+ "epoch": 1.04,
841
+ "learning_rate": 3.269853124222057e-05,
842
+ "loss": 1.6811,
843
+ "step": 1390
844
+ },
845
+ {
846
+ "epoch": 1.05,
847
+ "learning_rate": 3.2574060243963156e-05,
848
+ "loss": 1.6632,
849
+ "step": 1400
850
+ },
851
+ {
852
+ "epoch": 1.05,
853
+ "learning_rate": 3.244958924570575e-05,
854
+ "loss": 1.6529,
855
+ "step": 1410
856
+ },
857
+ {
858
+ "epoch": 1.06,
859
+ "learning_rate": 3.2325118247448345e-05,
860
+ "loss": 1.6943,
861
+ "step": 1420
862
+ },
863
+ {
864
+ "epoch": 1.07,
865
+ "learning_rate": 3.220064724919094e-05,
866
+ "loss": 1.6932,
867
+ "step": 1430
868
+ },
869
+ {
870
+ "epoch": 1.08,
871
+ "learning_rate": 3.2076176250933535e-05,
872
+ "loss": 1.6711,
873
+ "step": 1440
874
+ },
875
+ {
876
+ "epoch": 1.08,
877
+ "learning_rate": 3.195170525267613e-05,
878
+ "loss": 1.6711,
879
+ "step": 1450
880
+ },
881
+ {
882
+ "epoch": 1.09,
883
+ "learning_rate": 3.182723425441872e-05,
884
+ "loss": 1.6678,
885
+ "step": 1460
886
+ },
887
+ {
888
+ "epoch": 1.1,
889
+ "learning_rate": 3.170276325616132e-05,
890
+ "loss": 1.7058,
891
+ "step": 1470
892
+ },
893
+ {
894
+ "epoch": 1.11,
895
+ "learning_rate": 3.157829225790391e-05,
896
+ "loss": 1.7017,
897
+ "step": 1480
898
+ },
899
+ {
900
+ "epoch": 1.11,
901
+ "learning_rate": 3.14538212596465e-05,
902
+ "loss": 1.6862,
903
+ "step": 1490
904
+ },
905
+ {
906
+ "epoch": 1.12,
907
+ "learning_rate": 3.13293502613891e-05,
908
+ "loss": 1.6425,
909
+ "step": 1500
910
+ },
911
+ {
912
+ "epoch": 1.13,
913
+ "learning_rate": 3.120487926313169e-05,
914
+ "loss": 1.6977,
915
+ "step": 1510
916
+ },
917
+ {
918
+ "epoch": 1.14,
919
+ "learning_rate": 3.108040826487428e-05,
920
+ "loss": 1.698,
921
+ "step": 1520
922
+ },
923
+ {
924
+ "epoch": 1.14,
925
+ "learning_rate": 3.095593726661688e-05,
926
+ "loss": 1.6718,
927
+ "step": 1530
928
+ },
929
+ {
930
+ "epoch": 1.15,
931
+ "learning_rate": 3.083146626835947e-05,
932
+ "loss": 1.7025,
933
+ "step": 1540
934
+ },
935
+ {
936
+ "epoch": 1.16,
937
+ "learning_rate": 3.0706995270102065e-05,
938
+ "loss": 1.665,
939
+ "step": 1550
940
+ },
941
+ {
942
+ "epoch": 1.17,
943
+ "learning_rate": 3.058252427184466e-05,
944
+ "loss": 1.7035,
945
+ "step": 1560
946
+ },
947
+ {
948
+ "epoch": 1.17,
949
+ "learning_rate": 3.0458053273587255e-05,
950
+ "loss": 1.6849,
951
+ "step": 1570
952
+ },
953
+ {
954
+ "epoch": 1.18,
955
+ "learning_rate": 3.0333582275329846e-05,
956
+ "loss": 1.6345,
957
+ "step": 1580
958
+ },
959
+ {
960
+ "epoch": 1.19,
961
+ "learning_rate": 3.0209111277072444e-05,
962
+ "loss": 1.6593,
963
+ "step": 1590
964
+ },
965
+ {
966
+ "epoch": 1.19,
967
+ "learning_rate": 3.0084640278815036e-05,
968
+ "loss": 1.6248,
969
+ "step": 1600
970
+ },
971
+ {
972
+ "epoch": 1.2,
973
+ "learning_rate": 2.996016928055763e-05,
974
+ "loss": 1.6827,
975
+ "step": 1610
976
+ },
977
+ {
978
+ "epoch": 1.21,
979
+ "learning_rate": 2.9835698282300222e-05,
980
+ "loss": 1.7054,
981
+ "step": 1620
982
+ },
983
+ {
984
+ "epoch": 1.22,
985
+ "learning_rate": 2.971122728404282e-05,
986
+ "loss": 1.6518,
987
+ "step": 1630
988
+ },
989
+ {
990
+ "epoch": 1.22,
991
+ "learning_rate": 2.9586756285785412e-05,
992
+ "loss": 1.6795,
993
+ "step": 1640
994
+ },
995
+ {
996
+ "epoch": 1.23,
997
+ "learning_rate": 2.9462285287528007e-05,
998
+ "loss": 1.6286,
999
+ "step": 1650
1000
+ },
1001
+ {
1002
+ "epoch": 1.24,
1003
+ "learning_rate": 2.93378142892706e-05,
1004
+ "loss": 1.6599,
1005
+ "step": 1660
1006
+ },
1007
+ {
1008
+ "epoch": 1.25,
1009
+ "learning_rate": 2.9213343291013197e-05,
1010
+ "loss": 1.6902,
1011
+ "step": 1670
1012
+ },
1013
+ {
1014
+ "epoch": 1.25,
1015
+ "learning_rate": 2.9088872292755788e-05,
1016
+ "loss": 1.6551,
1017
+ "step": 1680
1018
+ },
1019
+ {
1020
+ "epoch": 1.26,
1021
+ "learning_rate": 2.8964401294498383e-05,
1022
+ "loss": 1.6493,
1023
+ "step": 1690
1024
+ },
1025
+ {
1026
+ "epoch": 1.27,
1027
+ "learning_rate": 2.8839930296240975e-05,
1028
+ "loss": 1.7094,
1029
+ "step": 1700
1030
+ },
1031
+ {
1032
+ "epoch": 1.28,
1033
+ "learning_rate": 2.8715459297983573e-05,
1034
+ "loss": 1.6714,
1035
+ "step": 1710
1036
+ },
1037
+ {
1038
+ "epoch": 1.28,
1039
+ "learning_rate": 2.859098829972616e-05,
1040
+ "loss": 1.6296,
1041
+ "step": 1720
1042
+ },
1043
+ {
1044
+ "epoch": 1.29,
1045
+ "learning_rate": 2.846651730146876e-05,
1046
+ "loss": 1.6882,
1047
+ "step": 1730
1048
+ },
1049
+ {
1050
+ "epoch": 1.3,
1051
+ "learning_rate": 2.834204630321135e-05,
1052
+ "loss": 1.6493,
1053
+ "step": 1740
1054
+ },
1055
+ {
1056
+ "epoch": 1.31,
1057
+ "learning_rate": 2.8217575304953945e-05,
1058
+ "loss": 1.6489,
1059
+ "step": 1750
1060
+ },
1061
+ {
1062
+ "epoch": 1.31,
1063
+ "learning_rate": 2.8093104306696544e-05,
1064
+ "loss": 1.6683,
1065
+ "step": 1760
1066
+ },
1067
+ {
1068
+ "epoch": 1.32,
1069
+ "learning_rate": 2.7968633308439135e-05,
1070
+ "loss": 1.6988,
1071
+ "step": 1770
1072
+ },
1073
+ {
1074
+ "epoch": 1.33,
1075
+ "learning_rate": 2.784416231018173e-05,
1076
+ "loss": 1.648,
1077
+ "step": 1780
1078
+ },
1079
+ {
1080
+ "epoch": 1.34,
1081
+ "learning_rate": 2.771969131192432e-05,
1082
+ "loss": 1.6655,
1083
+ "step": 1790
1084
+ },
1085
+ {
1086
+ "epoch": 1.34,
1087
+ "learning_rate": 2.759522031366692e-05,
1088
+ "loss": 1.6709,
1089
+ "step": 1800
1090
+ },
1091
+ {
1092
+ "epoch": 1.35,
1093
+ "learning_rate": 2.747074931540951e-05,
1094
+ "loss": 1.665,
1095
+ "step": 1810
1096
+ },
1097
+ {
1098
+ "epoch": 1.36,
1099
+ "learning_rate": 2.7346278317152106e-05,
1100
+ "loss": 1.6411,
1101
+ "step": 1820
1102
+ },
1103
+ {
1104
+ "epoch": 1.37,
1105
+ "learning_rate": 2.7221807318894698e-05,
1106
+ "loss": 1.6913,
1107
+ "step": 1830
1108
+ },
1109
+ {
1110
+ "epoch": 1.37,
1111
+ "learning_rate": 2.7097336320637296e-05,
1112
+ "loss": 1.6906,
1113
+ "step": 1840
1114
+ },
1115
+ {
1116
+ "epoch": 1.38,
1117
+ "learning_rate": 2.6972865322379887e-05,
1118
+ "loss": 1.6345,
1119
+ "step": 1850
1120
+ },
1121
+ {
1122
+ "epoch": 1.39,
1123
+ "learning_rate": 2.6848394324122482e-05,
1124
+ "loss": 1.6903,
1125
+ "step": 1860
1126
+ },
1127
+ {
1128
+ "epoch": 1.4,
1129
+ "learning_rate": 2.6723923325865074e-05,
1130
+ "loss": 1.6819,
1131
+ "step": 1870
1132
+ },
1133
+ {
1134
+ "epoch": 1.4,
1135
+ "learning_rate": 2.6599452327607672e-05,
1136
+ "loss": 1.6393,
1137
+ "step": 1880
1138
+ },
1139
+ {
1140
+ "epoch": 1.41,
1141
+ "learning_rate": 2.6474981329350264e-05,
1142
+ "loss": 1.6725,
1143
+ "step": 1890
1144
+ },
1145
+ {
1146
+ "epoch": 1.42,
1147
+ "learning_rate": 2.635051033109286e-05,
1148
+ "loss": 1.6736,
1149
+ "step": 1900
1150
+ },
1151
+ {
1152
+ "epoch": 1.43,
1153
+ "learning_rate": 2.622603933283545e-05,
1154
+ "loss": 1.6734,
1155
+ "step": 1910
1156
+ },
1157
+ {
1158
+ "epoch": 1.43,
1159
+ "learning_rate": 2.6101568334578048e-05,
1160
+ "loss": 1.6761,
1161
+ "step": 1920
1162
+ },
1163
+ {
1164
+ "epoch": 1.44,
1165
+ "learning_rate": 2.597709733632064e-05,
1166
+ "loss": 1.6842,
1167
+ "step": 1930
1168
+ },
1169
+ {
1170
+ "epoch": 1.45,
1171
+ "learning_rate": 2.5852626338063234e-05,
1172
+ "loss": 1.7064,
1173
+ "step": 1940
1174
+ },
1175
+ {
1176
+ "epoch": 1.46,
1177
+ "learning_rate": 2.5728155339805826e-05,
1178
+ "loss": 1.7036,
1179
+ "step": 1950
1180
+ },
1181
+ {
1182
+ "epoch": 1.46,
1183
+ "learning_rate": 2.5603684341548424e-05,
1184
+ "loss": 1.6563,
1185
+ "step": 1960
1186
+ },
1187
+ {
1188
+ "epoch": 1.47,
1189
+ "learning_rate": 2.5479213343291016e-05,
1190
+ "loss": 1.6796,
1191
+ "step": 1970
1192
+ },
1193
+ {
1194
+ "epoch": 1.48,
1195
+ "learning_rate": 2.535474234503361e-05,
1196
+ "loss": 1.6469,
1197
+ "step": 1980
1198
+ },
1199
+ {
1200
+ "epoch": 1.49,
1201
+ "learning_rate": 2.5230271346776202e-05,
1202
+ "loss": 1.6681,
1203
+ "step": 1990
1204
+ },
1205
+ {
1206
+ "epoch": 1.49,
1207
+ "learning_rate": 2.51058003485188e-05,
1208
+ "loss": 1.6632,
1209
+ "step": 2000
1210
+ }
1211
+ ],
1212
+ "logging_steps": 10,
1213
+ "max_steps": 4017,
1214
+ "num_input_tokens_seen": 0,
1215
+ "num_train_epochs": 3,
1216
+ "save_steps": 500,
1217
+ "total_flos": 1.26404964384768e+18,
1218
+ "train_batch_size": 4,
1219
+ "trial_name": null,
1220
+ "trial_params": null
1221
+ }
119_Llama-2-13b-hf/checkpoint-2000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:95fdd19e2ed6f33177f737bfa3ccdb7380f84c750beb151600638af981e7df02
3
+ size 4283
119_Llama-2-13b-hf/checkpoint-2500/README.md ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+ ### Framework versions
7
+
8
+
9
+ - PEFT 0.4.0
119_Llama-2-13b-hf/checkpoint-2500/adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "meta-llama/Llama-2-13b-hf",
4
+ "bias": "none",
5
+ "fan_in_fan_out": false,
6
+ "inference_mode": true,
7
+ "init_lora_weights": true,
8
+ "layers_pattern": null,
9
+ "layers_to_transform": null,
10
+ "lora_alpha": 32,
11
+ "lora_dropout": 0.05,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 8,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "q_proj",
18
+ "v_proj"
19
+ ],
20
+ "task_type": "CAUSAL_LM"
21
+ }
119_Llama-2-13b-hf/checkpoint-2500/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:796b34690020b48717a922b108f964fccf6ab2f0a6f17407947276709c19d949
3
+ size 26235704
119_Llama-2-13b-hf/checkpoint-2500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f0c586887948d6cd63d728376d9f8b381adcb0da8771088a2b86ddcedc8c68cf
3
+ size 52567237
119_Llama-2-13b-hf/checkpoint-2500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:183a9cb64a9222b02973fc413a589dae4a9138b9c0e04ce38d1397ef4acd70a0
3
+ size 14575
119_Llama-2-13b-hf/checkpoint-2500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:16392c3406dc703a482a2728295444e19fbb823433cc4ac55c8e057f0228867a
3
+ size 627
119_Llama-2-13b-hf/checkpoint-2500/trainer_state.json ADDED
@@ -0,0 +1,1521 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.8670649738610905,
5
+ "eval_steps": 500,
6
+ "global_step": 2500,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.01,
13
+ "learning_rate": 4.98755290017426e-05,
14
+ "loss": 1.7891,
15
+ "step": 10
16
+ },
17
+ {
18
+ "epoch": 0.01,
19
+ "learning_rate": 4.975105800348519e-05,
20
+ "loss": 1.7869,
21
+ "step": 20
22
+ },
23
+ {
24
+ "epoch": 0.02,
25
+ "learning_rate": 4.962658700522779e-05,
26
+ "loss": 1.7868,
27
+ "step": 30
28
+ },
29
+ {
30
+ "epoch": 0.03,
31
+ "learning_rate": 4.9502116006970375e-05,
32
+ "loss": 1.7547,
33
+ "step": 40
34
+ },
35
+ {
36
+ "epoch": 0.04,
37
+ "learning_rate": 4.937764500871298e-05,
38
+ "loss": 1.6772,
39
+ "step": 50
40
+ },
41
+ {
42
+ "epoch": 0.04,
43
+ "learning_rate": 4.9253174010455565e-05,
44
+ "loss": 1.7076,
45
+ "step": 60
46
+ },
47
+ {
48
+ "epoch": 0.05,
49
+ "learning_rate": 4.912870301219816e-05,
50
+ "loss": 1.6948,
51
+ "step": 70
52
+ },
53
+ {
54
+ "epoch": 0.06,
55
+ "learning_rate": 4.9004232013940755e-05,
56
+ "loss": 1.7135,
57
+ "step": 80
58
+ },
59
+ {
60
+ "epoch": 0.07,
61
+ "learning_rate": 4.887976101568335e-05,
62
+ "loss": 1.7007,
63
+ "step": 90
64
+ },
65
+ {
66
+ "epoch": 0.07,
67
+ "learning_rate": 4.8755290017425944e-05,
68
+ "loss": 1.7247,
69
+ "step": 100
70
+ },
71
+ {
72
+ "epoch": 0.08,
73
+ "learning_rate": 4.863081901916854e-05,
74
+ "loss": 1.719,
75
+ "step": 110
76
+ },
77
+ {
78
+ "epoch": 0.09,
79
+ "learning_rate": 4.850634802091113e-05,
80
+ "loss": 1.7209,
81
+ "step": 120
82
+ },
83
+ {
84
+ "epoch": 0.1,
85
+ "learning_rate": 4.838187702265373e-05,
86
+ "loss": 1.7354,
87
+ "step": 130
88
+ },
89
+ {
90
+ "epoch": 0.1,
91
+ "learning_rate": 4.825740602439632e-05,
92
+ "loss": 1.7025,
93
+ "step": 140
94
+ },
95
+ {
96
+ "epoch": 0.11,
97
+ "learning_rate": 4.813293502613891e-05,
98
+ "loss": 1.6664,
99
+ "step": 150
100
+ },
101
+ {
102
+ "epoch": 0.12,
103
+ "learning_rate": 4.800846402788151e-05,
104
+ "loss": 1.6851,
105
+ "step": 160
106
+ },
107
+ {
108
+ "epoch": 0.13,
109
+ "learning_rate": 4.78839930296241e-05,
110
+ "loss": 1.6749,
111
+ "step": 170
112
+ },
113
+ {
114
+ "epoch": 0.13,
115
+ "learning_rate": 4.7759522031366697e-05,
116
+ "loss": 1.6887,
117
+ "step": 180
118
+ },
119
+ {
120
+ "epoch": 0.14,
121
+ "learning_rate": 4.763505103310929e-05,
122
+ "loss": 1.6571,
123
+ "step": 190
124
+ },
125
+ {
126
+ "epoch": 0.15,
127
+ "learning_rate": 4.751058003485188e-05,
128
+ "loss": 1.7136,
129
+ "step": 200
130
+ },
131
+ {
132
+ "epoch": 0.16,
133
+ "learning_rate": 4.738610903659448e-05,
134
+ "loss": 1.7098,
135
+ "step": 210
136
+ },
137
+ {
138
+ "epoch": 0.16,
139
+ "learning_rate": 4.726163803833707e-05,
140
+ "loss": 1.7285,
141
+ "step": 220
142
+ },
143
+ {
144
+ "epoch": 0.17,
145
+ "learning_rate": 4.7137167040079664e-05,
146
+ "loss": 1.7038,
147
+ "step": 230
148
+ },
149
+ {
150
+ "epoch": 0.18,
151
+ "learning_rate": 4.701269604182226e-05,
152
+ "loss": 1.692,
153
+ "step": 240
154
+ },
155
+ {
156
+ "epoch": 0.19,
157
+ "learning_rate": 4.6888225043564854e-05,
158
+ "loss": 1.6825,
159
+ "step": 250
160
+ },
161
+ {
162
+ "epoch": 0.19,
163
+ "learning_rate": 4.676375404530744e-05,
164
+ "loss": 1.7254,
165
+ "step": 260
166
+ },
167
+ {
168
+ "epoch": 0.2,
169
+ "learning_rate": 4.6639283047050044e-05,
170
+ "loss": 1.6765,
171
+ "step": 270
172
+ },
173
+ {
174
+ "epoch": 0.21,
175
+ "learning_rate": 4.651481204879263e-05,
176
+ "loss": 1.6796,
177
+ "step": 280
178
+ },
179
+ {
180
+ "epoch": 0.22,
181
+ "learning_rate": 4.6390341050535227e-05,
182
+ "loss": 1.7302,
183
+ "step": 290
184
+ },
185
+ {
186
+ "epoch": 0.22,
187
+ "learning_rate": 4.626587005227782e-05,
188
+ "loss": 1.7469,
189
+ "step": 300
190
+ },
191
+ {
192
+ "epoch": 0.23,
193
+ "learning_rate": 4.6141399054020416e-05,
194
+ "loss": 1.6345,
195
+ "step": 310
196
+ },
197
+ {
198
+ "epoch": 0.24,
199
+ "learning_rate": 4.601692805576301e-05,
200
+ "loss": 1.6761,
201
+ "step": 320
202
+ },
203
+ {
204
+ "epoch": 0.25,
205
+ "learning_rate": 4.5892457057505606e-05,
206
+ "loss": 1.6602,
207
+ "step": 330
208
+ },
209
+ {
210
+ "epoch": 0.25,
211
+ "learning_rate": 4.5767986059248194e-05,
212
+ "loss": 1.6637,
213
+ "step": 340
214
+ },
215
+ {
216
+ "epoch": 0.26,
217
+ "learning_rate": 4.5643515060990796e-05,
218
+ "loss": 1.6953,
219
+ "step": 350
220
+ },
221
+ {
222
+ "epoch": 0.27,
223
+ "learning_rate": 4.5519044062733384e-05,
224
+ "loss": 1.6829,
225
+ "step": 360
226
+ },
227
+ {
228
+ "epoch": 0.28,
229
+ "learning_rate": 4.539457306447598e-05,
230
+ "loss": 1.657,
231
+ "step": 370
232
+ },
233
+ {
234
+ "epoch": 0.28,
235
+ "learning_rate": 4.5270102066218574e-05,
236
+ "loss": 1.647,
237
+ "step": 380
238
+ },
239
+ {
240
+ "epoch": 0.29,
241
+ "learning_rate": 4.514563106796117e-05,
242
+ "loss": 1.6854,
243
+ "step": 390
244
+ },
245
+ {
246
+ "epoch": 0.3,
247
+ "learning_rate": 4.502116006970376e-05,
248
+ "loss": 1.6737,
249
+ "step": 400
250
+ },
251
+ {
252
+ "epoch": 0.31,
253
+ "learning_rate": 4.489668907144636e-05,
254
+ "loss": 1.6938,
255
+ "step": 410
256
+ },
257
+ {
258
+ "epoch": 0.31,
259
+ "learning_rate": 4.4772218073188946e-05,
260
+ "loss": 1.6721,
261
+ "step": 420
262
+ },
263
+ {
264
+ "epoch": 0.32,
265
+ "learning_rate": 4.464774707493155e-05,
266
+ "loss": 1.6683,
267
+ "step": 430
268
+ },
269
+ {
270
+ "epoch": 0.33,
271
+ "learning_rate": 4.4523276076674136e-05,
272
+ "loss": 1.6229,
273
+ "step": 440
274
+ },
275
+ {
276
+ "epoch": 0.34,
277
+ "learning_rate": 4.439880507841673e-05,
278
+ "loss": 1.6593,
279
+ "step": 450
280
+ },
281
+ {
282
+ "epoch": 0.34,
283
+ "learning_rate": 4.4274334080159326e-05,
284
+ "loss": 1.6844,
285
+ "step": 460
286
+ },
287
+ {
288
+ "epoch": 0.35,
289
+ "learning_rate": 4.414986308190192e-05,
290
+ "loss": 1.6754,
291
+ "step": 470
292
+ },
293
+ {
294
+ "epoch": 0.36,
295
+ "learning_rate": 4.4025392083644516e-05,
296
+ "loss": 1.6747,
297
+ "step": 480
298
+ },
299
+ {
300
+ "epoch": 0.37,
301
+ "learning_rate": 4.390092108538711e-05,
302
+ "loss": 1.6789,
303
+ "step": 490
304
+ },
305
+ {
306
+ "epoch": 0.37,
307
+ "learning_rate": 4.37764500871297e-05,
308
+ "loss": 1.6956,
309
+ "step": 500
310
+ },
311
+ {
312
+ "epoch": 0.38,
313
+ "learning_rate": 4.36519790888723e-05,
314
+ "loss": 1.689,
315
+ "step": 510
316
+ },
317
+ {
318
+ "epoch": 0.39,
319
+ "learning_rate": 4.352750809061489e-05,
320
+ "loss": 1.673,
321
+ "step": 520
322
+ },
323
+ {
324
+ "epoch": 0.4,
325
+ "learning_rate": 4.340303709235748e-05,
326
+ "loss": 1.6968,
327
+ "step": 530
328
+ },
329
+ {
330
+ "epoch": 0.4,
331
+ "learning_rate": 4.327856609410008e-05,
332
+ "loss": 1.6569,
333
+ "step": 540
334
+ },
335
+ {
336
+ "epoch": 0.41,
337
+ "learning_rate": 4.315409509584267e-05,
338
+ "loss": 1.6777,
339
+ "step": 550
340
+ },
341
+ {
342
+ "epoch": 0.42,
343
+ "learning_rate": 4.302962409758526e-05,
344
+ "loss": 1.71,
345
+ "step": 560
346
+ },
347
+ {
348
+ "epoch": 0.43,
349
+ "learning_rate": 4.290515309932786e-05,
350
+ "loss": 1.6943,
351
+ "step": 570
352
+ },
353
+ {
354
+ "epoch": 0.43,
355
+ "learning_rate": 4.278068210107045e-05,
356
+ "loss": 1.6925,
357
+ "step": 580
358
+ },
359
+ {
360
+ "epoch": 0.44,
361
+ "learning_rate": 4.2656211102813046e-05,
362
+ "loss": 1.6628,
363
+ "step": 590
364
+ },
365
+ {
366
+ "epoch": 0.45,
367
+ "learning_rate": 4.253174010455564e-05,
368
+ "loss": 1.7123,
369
+ "step": 600
370
+ },
371
+ {
372
+ "epoch": 0.46,
373
+ "learning_rate": 4.2407269106298235e-05,
374
+ "loss": 1.6772,
375
+ "step": 610
376
+ },
377
+ {
378
+ "epoch": 0.46,
379
+ "learning_rate": 4.228279810804083e-05,
380
+ "loss": 1.6429,
381
+ "step": 620
382
+ },
383
+ {
384
+ "epoch": 0.47,
385
+ "learning_rate": 4.2158327109783425e-05,
386
+ "loss": 1.6472,
387
+ "step": 630
388
+ },
389
+ {
390
+ "epoch": 0.48,
391
+ "learning_rate": 4.203385611152601e-05,
392
+ "loss": 1.6825,
393
+ "step": 640
394
+ },
395
+ {
396
+ "epoch": 0.49,
397
+ "learning_rate": 4.1909385113268615e-05,
398
+ "loss": 1.6825,
399
+ "step": 650
400
+ },
401
+ {
402
+ "epoch": 0.49,
403
+ "learning_rate": 4.17849141150112e-05,
404
+ "loss": 1.6818,
405
+ "step": 660
406
+ },
407
+ {
408
+ "epoch": 0.5,
409
+ "learning_rate": 4.16604431167538e-05,
410
+ "loss": 1.7325,
411
+ "step": 670
412
+ },
413
+ {
414
+ "epoch": 0.51,
415
+ "learning_rate": 4.153597211849639e-05,
416
+ "loss": 1.7008,
417
+ "step": 680
418
+ },
419
+ {
420
+ "epoch": 0.52,
421
+ "learning_rate": 4.141150112023899e-05,
422
+ "loss": 1.655,
423
+ "step": 690
424
+ },
425
+ {
426
+ "epoch": 0.52,
427
+ "learning_rate": 4.128703012198158e-05,
428
+ "loss": 1.6718,
429
+ "step": 700
430
+ },
431
+ {
432
+ "epoch": 0.53,
433
+ "learning_rate": 4.116255912372418e-05,
434
+ "loss": 1.6412,
435
+ "step": 710
436
+ },
437
+ {
438
+ "epoch": 0.54,
439
+ "learning_rate": 4.1038088125466765e-05,
440
+ "loss": 1.7091,
441
+ "step": 720
442
+ },
443
+ {
444
+ "epoch": 0.55,
445
+ "learning_rate": 4.091361712720937e-05,
446
+ "loss": 1.726,
447
+ "step": 730
448
+ },
449
+ {
450
+ "epoch": 0.55,
451
+ "learning_rate": 4.0789146128951955e-05,
452
+ "loss": 1.6655,
453
+ "step": 740
454
+ },
455
+ {
456
+ "epoch": 0.56,
457
+ "learning_rate": 4.066467513069455e-05,
458
+ "loss": 1.7242,
459
+ "step": 750
460
+ },
461
+ {
462
+ "epoch": 0.57,
463
+ "learning_rate": 4.0540204132437145e-05,
464
+ "loss": 1.6714,
465
+ "step": 760
466
+ },
467
+ {
468
+ "epoch": 0.58,
469
+ "learning_rate": 4.041573313417974e-05,
470
+ "loss": 1.6646,
471
+ "step": 770
472
+ },
473
+ {
474
+ "epoch": 0.58,
475
+ "learning_rate": 4.029126213592233e-05,
476
+ "loss": 1.6809,
477
+ "step": 780
478
+ },
479
+ {
480
+ "epoch": 0.59,
481
+ "learning_rate": 4.016679113766493e-05,
482
+ "loss": 1.6868,
483
+ "step": 790
484
+ },
485
+ {
486
+ "epoch": 0.6,
487
+ "learning_rate": 4.004232013940752e-05,
488
+ "loss": 1.694,
489
+ "step": 800
490
+ },
491
+ {
492
+ "epoch": 0.6,
493
+ "learning_rate": 3.991784914115011e-05,
494
+ "loss": 1.6801,
495
+ "step": 810
496
+ },
497
+ {
498
+ "epoch": 0.61,
499
+ "learning_rate": 3.979337814289271e-05,
500
+ "loss": 1.6861,
501
+ "step": 820
502
+ },
503
+ {
504
+ "epoch": 0.62,
505
+ "learning_rate": 3.96689071446353e-05,
506
+ "loss": 1.6646,
507
+ "step": 830
508
+ },
509
+ {
510
+ "epoch": 0.63,
511
+ "learning_rate": 3.95444361463779e-05,
512
+ "loss": 1.6764,
513
+ "step": 840
514
+ },
515
+ {
516
+ "epoch": 0.63,
517
+ "learning_rate": 3.941996514812049e-05,
518
+ "loss": 1.6821,
519
+ "step": 850
520
+ },
521
+ {
522
+ "epoch": 0.64,
523
+ "learning_rate": 3.929549414986308e-05,
524
+ "loss": 1.6681,
525
+ "step": 860
526
+ },
527
+ {
528
+ "epoch": 0.65,
529
+ "learning_rate": 3.917102315160568e-05,
530
+ "loss": 1.6847,
531
+ "step": 870
532
+ },
533
+ {
534
+ "epoch": 0.66,
535
+ "learning_rate": 3.904655215334827e-05,
536
+ "loss": 1.6084,
537
+ "step": 880
538
+ },
539
+ {
540
+ "epoch": 0.66,
541
+ "learning_rate": 3.8922081155090865e-05,
542
+ "loss": 1.6624,
543
+ "step": 890
544
+ },
545
+ {
546
+ "epoch": 0.67,
547
+ "learning_rate": 3.879761015683346e-05,
548
+ "loss": 1.6619,
549
+ "step": 900
550
+ },
551
+ {
552
+ "epoch": 0.68,
553
+ "learning_rate": 3.8673139158576054e-05,
554
+ "loss": 1.7024,
555
+ "step": 910
556
+ },
557
+ {
558
+ "epoch": 0.69,
559
+ "learning_rate": 3.854866816031865e-05,
560
+ "loss": 1.7301,
561
+ "step": 920
562
+ },
563
+ {
564
+ "epoch": 0.69,
565
+ "learning_rate": 3.8424197162061244e-05,
566
+ "loss": 1.6907,
567
+ "step": 930
568
+ },
569
+ {
570
+ "epoch": 0.7,
571
+ "learning_rate": 3.829972616380383e-05,
572
+ "loss": 1.7046,
573
+ "step": 940
574
+ },
575
+ {
576
+ "epoch": 0.71,
577
+ "learning_rate": 3.8175255165546434e-05,
578
+ "loss": 1.6661,
579
+ "step": 950
580
+ },
581
+ {
582
+ "epoch": 0.72,
583
+ "learning_rate": 3.805078416728902e-05,
584
+ "loss": 1.6981,
585
+ "step": 960
586
+ },
587
+ {
588
+ "epoch": 0.72,
589
+ "learning_rate": 3.792631316903162e-05,
590
+ "loss": 1.662,
591
+ "step": 970
592
+ },
593
+ {
594
+ "epoch": 0.73,
595
+ "learning_rate": 3.780184217077421e-05,
596
+ "loss": 1.7073,
597
+ "step": 980
598
+ },
599
+ {
600
+ "epoch": 0.74,
601
+ "learning_rate": 3.7677371172516806e-05,
602
+ "loss": 1.6488,
603
+ "step": 990
604
+ },
605
+ {
606
+ "epoch": 0.75,
607
+ "learning_rate": 3.7552900174259395e-05,
608
+ "loss": 1.6708,
609
+ "step": 1000
610
+ },
611
+ {
612
+ "epoch": 0.75,
613
+ "learning_rate": 3.7428429176001996e-05,
614
+ "loss": 1.7045,
615
+ "step": 1010
616
+ },
617
+ {
618
+ "epoch": 0.76,
619
+ "learning_rate": 3.7303958177744584e-05,
620
+ "loss": 1.6842,
621
+ "step": 1020
622
+ },
623
+ {
624
+ "epoch": 0.77,
625
+ "learning_rate": 3.717948717948718e-05,
626
+ "loss": 1.6682,
627
+ "step": 1030
628
+ },
629
+ {
630
+ "epoch": 0.78,
631
+ "learning_rate": 3.7055016181229774e-05,
632
+ "loss": 1.636,
633
+ "step": 1040
634
+ },
635
+ {
636
+ "epoch": 0.78,
637
+ "learning_rate": 3.693054518297237e-05,
638
+ "loss": 1.6801,
639
+ "step": 1050
640
+ },
641
+ {
642
+ "epoch": 0.79,
643
+ "learning_rate": 3.6806074184714964e-05,
644
+ "loss": 1.6371,
645
+ "step": 1060
646
+ },
647
+ {
648
+ "epoch": 0.8,
649
+ "learning_rate": 3.668160318645756e-05,
650
+ "loss": 1.6755,
651
+ "step": 1070
652
+ },
653
+ {
654
+ "epoch": 0.81,
655
+ "learning_rate": 3.655713218820015e-05,
656
+ "loss": 1.7046,
657
+ "step": 1080
658
+ },
659
+ {
660
+ "epoch": 0.81,
661
+ "learning_rate": 3.643266118994275e-05,
662
+ "loss": 1.707,
663
+ "step": 1090
664
+ },
665
+ {
666
+ "epoch": 0.82,
667
+ "learning_rate": 3.6308190191685337e-05,
668
+ "loss": 1.6568,
669
+ "step": 1100
670
+ },
671
+ {
672
+ "epoch": 0.83,
673
+ "learning_rate": 3.618371919342793e-05,
674
+ "loss": 1.6447,
675
+ "step": 1110
676
+ },
677
+ {
678
+ "epoch": 0.84,
679
+ "learning_rate": 3.6059248195170526e-05,
680
+ "loss": 1.691,
681
+ "step": 1120
682
+ },
683
+ {
684
+ "epoch": 0.84,
685
+ "learning_rate": 3.593477719691312e-05,
686
+ "loss": 1.6747,
687
+ "step": 1130
688
+ },
689
+ {
690
+ "epoch": 0.85,
691
+ "learning_rate": 3.5810306198655716e-05,
692
+ "loss": 1.656,
693
+ "step": 1140
694
+ },
695
+ {
696
+ "epoch": 0.86,
697
+ "learning_rate": 3.568583520039831e-05,
698
+ "loss": 1.6485,
699
+ "step": 1150
700
+ },
701
+ {
702
+ "epoch": 0.87,
703
+ "learning_rate": 3.55613642021409e-05,
704
+ "loss": 1.642,
705
+ "step": 1160
706
+ },
707
+ {
708
+ "epoch": 0.87,
709
+ "learning_rate": 3.54368932038835e-05,
710
+ "loss": 1.6891,
711
+ "step": 1170
712
+ },
713
+ {
714
+ "epoch": 0.88,
715
+ "learning_rate": 3.531242220562609e-05,
716
+ "loss": 1.716,
717
+ "step": 1180
718
+ },
719
+ {
720
+ "epoch": 0.89,
721
+ "learning_rate": 3.5187951207368684e-05,
722
+ "loss": 1.6785,
723
+ "step": 1190
724
+ },
725
+ {
726
+ "epoch": 0.9,
727
+ "learning_rate": 3.506348020911128e-05,
728
+ "loss": 1.7053,
729
+ "step": 1200
730
+ },
731
+ {
732
+ "epoch": 0.9,
733
+ "learning_rate": 3.493900921085387e-05,
734
+ "loss": 1.6833,
735
+ "step": 1210
736
+ },
737
+ {
738
+ "epoch": 0.91,
739
+ "learning_rate": 3.481453821259646e-05,
740
+ "loss": 1.7336,
741
+ "step": 1220
742
+ },
743
+ {
744
+ "epoch": 0.92,
745
+ "learning_rate": 3.469006721433906e-05,
746
+ "loss": 1.6496,
747
+ "step": 1230
748
+ },
749
+ {
750
+ "epoch": 0.93,
751
+ "learning_rate": 3.456559621608165e-05,
752
+ "loss": 1.7014,
753
+ "step": 1240
754
+ },
755
+ {
756
+ "epoch": 0.93,
757
+ "learning_rate": 3.4441125217824246e-05,
758
+ "loss": 1.6502,
759
+ "step": 1250
760
+ },
761
+ {
762
+ "epoch": 0.94,
763
+ "learning_rate": 3.431665421956684e-05,
764
+ "loss": 1.6617,
765
+ "step": 1260
766
+ },
767
+ {
768
+ "epoch": 0.95,
769
+ "learning_rate": 3.4192183221309436e-05,
770
+ "loss": 1.7155,
771
+ "step": 1270
772
+ },
773
+ {
774
+ "epoch": 0.96,
775
+ "learning_rate": 3.406771222305203e-05,
776
+ "loss": 1.691,
777
+ "step": 1280
778
+ },
779
+ {
780
+ "epoch": 0.96,
781
+ "learning_rate": 3.3943241224794625e-05,
782
+ "loss": 1.6601,
783
+ "step": 1290
784
+ },
785
+ {
786
+ "epoch": 0.97,
787
+ "learning_rate": 3.3818770226537214e-05,
788
+ "loss": 1.6607,
789
+ "step": 1300
790
+ },
791
+ {
792
+ "epoch": 0.98,
793
+ "learning_rate": 3.3694299228279815e-05,
794
+ "loss": 1.7053,
795
+ "step": 1310
796
+ },
797
+ {
798
+ "epoch": 0.99,
799
+ "learning_rate": 3.35698282300224e-05,
800
+ "loss": 1.6681,
801
+ "step": 1320
802
+ },
803
+ {
804
+ "epoch": 0.99,
805
+ "learning_rate": 3.3445357231765e-05,
806
+ "loss": 1.6712,
807
+ "step": 1330
808
+ },
809
+ {
810
+ "epoch": 1.0,
811
+ "learning_rate": 3.332088623350759e-05,
812
+ "loss": 1.6601,
813
+ "step": 1340
814
+ },
815
+ {
816
+ "epoch": 1.01,
817
+ "learning_rate": 3.319641523525019e-05,
818
+ "loss": 1.6522,
819
+ "step": 1350
820
+ },
821
+ {
822
+ "epoch": 1.02,
823
+ "learning_rate": 3.307194423699278e-05,
824
+ "loss": 1.6621,
825
+ "step": 1360
826
+ },
827
+ {
828
+ "epoch": 1.02,
829
+ "learning_rate": 3.294747323873538e-05,
830
+ "loss": 1.6651,
831
+ "step": 1370
832
+ },
833
+ {
834
+ "epoch": 1.03,
835
+ "learning_rate": 3.2823002240477966e-05,
836
+ "loss": 1.6665,
837
+ "step": 1380
838
+ },
839
+ {
840
+ "epoch": 1.04,
841
+ "learning_rate": 3.269853124222057e-05,
842
+ "loss": 1.6811,
843
+ "step": 1390
844
+ },
845
+ {
846
+ "epoch": 1.05,
847
+ "learning_rate": 3.2574060243963156e-05,
848
+ "loss": 1.6632,
849
+ "step": 1400
850
+ },
851
+ {
852
+ "epoch": 1.05,
853
+ "learning_rate": 3.244958924570575e-05,
854
+ "loss": 1.6529,
855
+ "step": 1410
856
+ },
857
+ {
858
+ "epoch": 1.06,
859
+ "learning_rate": 3.2325118247448345e-05,
860
+ "loss": 1.6943,
861
+ "step": 1420
862
+ },
863
+ {
864
+ "epoch": 1.07,
865
+ "learning_rate": 3.220064724919094e-05,
866
+ "loss": 1.6932,
867
+ "step": 1430
868
+ },
869
+ {
870
+ "epoch": 1.08,
871
+ "learning_rate": 3.2076176250933535e-05,
872
+ "loss": 1.6711,
873
+ "step": 1440
874
+ },
875
+ {
876
+ "epoch": 1.08,
877
+ "learning_rate": 3.195170525267613e-05,
878
+ "loss": 1.6711,
879
+ "step": 1450
880
+ },
881
+ {
882
+ "epoch": 1.09,
883
+ "learning_rate": 3.182723425441872e-05,
884
+ "loss": 1.6678,
885
+ "step": 1460
886
+ },
887
+ {
888
+ "epoch": 1.1,
889
+ "learning_rate": 3.170276325616132e-05,
890
+ "loss": 1.7058,
891
+ "step": 1470
892
+ },
893
+ {
894
+ "epoch": 1.11,
895
+ "learning_rate": 3.157829225790391e-05,
896
+ "loss": 1.7017,
897
+ "step": 1480
898
+ },
899
+ {
900
+ "epoch": 1.11,
901
+ "learning_rate": 3.14538212596465e-05,
902
+ "loss": 1.6862,
903
+ "step": 1490
904
+ },
905
+ {
906
+ "epoch": 1.12,
907
+ "learning_rate": 3.13293502613891e-05,
908
+ "loss": 1.6425,
909
+ "step": 1500
910
+ },
911
+ {
912
+ "epoch": 1.13,
913
+ "learning_rate": 3.120487926313169e-05,
914
+ "loss": 1.6977,
915
+ "step": 1510
916
+ },
917
+ {
918
+ "epoch": 1.14,
919
+ "learning_rate": 3.108040826487428e-05,
920
+ "loss": 1.698,
921
+ "step": 1520
922
+ },
923
+ {
924
+ "epoch": 1.14,
925
+ "learning_rate": 3.095593726661688e-05,
926
+ "loss": 1.6718,
927
+ "step": 1530
928
+ },
929
+ {
930
+ "epoch": 1.15,
931
+ "learning_rate": 3.083146626835947e-05,
932
+ "loss": 1.7025,
933
+ "step": 1540
934
+ },
935
+ {
936
+ "epoch": 1.16,
937
+ "learning_rate": 3.0706995270102065e-05,
938
+ "loss": 1.665,
939
+ "step": 1550
940
+ },
941
+ {
942
+ "epoch": 1.17,
943
+ "learning_rate": 3.058252427184466e-05,
944
+ "loss": 1.7035,
945
+ "step": 1560
946
+ },
947
+ {
948
+ "epoch": 1.17,
949
+ "learning_rate": 3.0458053273587255e-05,
950
+ "loss": 1.6849,
951
+ "step": 1570
952
+ },
953
+ {
954
+ "epoch": 1.18,
955
+ "learning_rate": 3.0333582275329846e-05,
956
+ "loss": 1.6345,
957
+ "step": 1580
958
+ },
959
+ {
960
+ "epoch": 1.19,
961
+ "learning_rate": 3.0209111277072444e-05,
962
+ "loss": 1.6593,
963
+ "step": 1590
964
+ },
965
+ {
966
+ "epoch": 1.19,
967
+ "learning_rate": 3.0084640278815036e-05,
968
+ "loss": 1.6248,
969
+ "step": 1600
970
+ },
971
+ {
972
+ "epoch": 1.2,
973
+ "learning_rate": 2.996016928055763e-05,
974
+ "loss": 1.6827,
975
+ "step": 1610
976
+ },
977
+ {
978
+ "epoch": 1.21,
979
+ "learning_rate": 2.9835698282300222e-05,
980
+ "loss": 1.7054,
981
+ "step": 1620
982
+ },
983
+ {
984
+ "epoch": 1.22,
985
+ "learning_rate": 2.971122728404282e-05,
986
+ "loss": 1.6518,
987
+ "step": 1630
988
+ },
989
+ {
990
+ "epoch": 1.22,
991
+ "learning_rate": 2.9586756285785412e-05,
992
+ "loss": 1.6795,
993
+ "step": 1640
994
+ },
995
+ {
996
+ "epoch": 1.23,
997
+ "learning_rate": 2.9462285287528007e-05,
998
+ "loss": 1.6286,
999
+ "step": 1650
1000
+ },
1001
+ {
1002
+ "epoch": 1.24,
1003
+ "learning_rate": 2.93378142892706e-05,
1004
+ "loss": 1.6599,
1005
+ "step": 1660
1006
+ },
1007
+ {
1008
+ "epoch": 1.25,
1009
+ "learning_rate": 2.9213343291013197e-05,
1010
+ "loss": 1.6902,
1011
+ "step": 1670
1012
+ },
1013
+ {
1014
+ "epoch": 1.25,
1015
+ "learning_rate": 2.9088872292755788e-05,
1016
+ "loss": 1.6551,
1017
+ "step": 1680
1018
+ },
1019
+ {
1020
+ "epoch": 1.26,
1021
+ "learning_rate": 2.8964401294498383e-05,
1022
+ "loss": 1.6493,
1023
+ "step": 1690
1024
+ },
1025
+ {
1026
+ "epoch": 1.27,
1027
+ "learning_rate": 2.8839930296240975e-05,
1028
+ "loss": 1.7094,
1029
+ "step": 1700
1030
+ },
1031
+ {
1032
+ "epoch": 1.28,
1033
+ "learning_rate": 2.8715459297983573e-05,
1034
+ "loss": 1.6714,
1035
+ "step": 1710
1036
+ },
1037
+ {
1038
+ "epoch": 1.28,
1039
+ "learning_rate": 2.859098829972616e-05,
1040
+ "loss": 1.6296,
1041
+ "step": 1720
1042
+ },
1043
+ {
1044
+ "epoch": 1.29,
1045
+ "learning_rate": 2.846651730146876e-05,
1046
+ "loss": 1.6882,
1047
+ "step": 1730
1048
+ },
1049
+ {
1050
+ "epoch": 1.3,
1051
+ "learning_rate": 2.834204630321135e-05,
1052
+ "loss": 1.6493,
1053
+ "step": 1740
1054
+ },
1055
+ {
1056
+ "epoch": 1.31,
1057
+ "learning_rate": 2.8217575304953945e-05,
1058
+ "loss": 1.6489,
1059
+ "step": 1750
1060
+ },
1061
+ {
1062
+ "epoch": 1.31,
1063
+ "learning_rate": 2.8093104306696544e-05,
1064
+ "loss": 1.6683,
1065
+ "step": 1760
1066
+ },
1067
+ {
1068
+ "epoch": 1.32,
1069
+ "learning_rate": 2.7968633308439135e-05,
1070
+ "loss": 1.6988,
1071
+ "step": 1770
1072
+ },
1073
+ {
1074
+ "epoch": 1.33,
1075
+ "learning_rate": 2.784416231018173e-05,
1076
+ "loss": 1.648,
1077
+ "step": 1780
1078
+ },
1079
+ {
1080
+ "epoch": 1.34,
1081
+ "learning_rate": 2.771969131192432e-05,
1082
+ "loss": 1.6655,
1083
+ "step": 1790
1084
+ },
1085
+ {
1086
+ "epoch": 1.34,
1087
+ "learning_rate": 2.759522031366692e-05,
1088
+ "loss": 1.6709,
1089
+ "step": 1800
1090
+ },
1091
+ {
1092
+ "epoch": 1.35,
1093
+ "learning_rate": 2.747074931540951e-05,
1094
+ "loss": 1.665,
1095
+ "step": 1810
1096
+ },
1097
+ {
1098
+ "epoch": 1.36,
1099
+ "learning_rate": 2.7346278317152106e-05,
1100
+ "loss": 1.6411,
1101
+ "step": 1820
1102
+ },
1103
+ {
1104
+ "epoch": 1.37,
1105
+ "learning_rate": 2.7221807318894698e-05,
1106
+ "loss": 1.6913,
1107
+ "step": 1830
1108
+ },
1109
+ {
1110
+ "epoch": 1.37,
1111
+ "learning_rate": 2.7097336320637296e-05,
1112
+ "loss": 1.6906,
1113
+ "step": 1840
1114
+ },
1115
+ {
1116
+ "epoch": 1.38,
1117
+ "learning_rate": 2.6972865322379887e-05,
1118
+ "loss": 1.6345,
1119
+ "step": 1850
1120
+ },
1121
+ {
1122
+ "epoch": 1.39,
1123
+ "learning_rate": 2.6848394324122482e-05,
1124
+ "loss": 1.6903,
1125
+ "step": 1860
1126
+ },
1127
+ {
1128
+ "epoch": 1.4,
1129
+ "learning_rate": 2.6723923325865074e-05,
1130
+ "loss": 1.6819,
1131
+ "step": 1870
1132
+ },
1133
+ {
1134
+ "epoch": 1.4,
1135
+ "learning_rate": 2.6599452327607672e-05,
1136
+ "loss": 1.6393,
1137
+ "step": 1880
1138
+ },
1139
+ {
1140
+ "epoch": 1.41,
1141
+ "learning_rate": 2.6474981329350264e-05,
1142
+ "loss": 1.6725,
1143
+ "step": 1890
1144
+ },
1145
+ {
1146
+ "epoch": 1.42,
1147
+ "learning_rate": 2.635051033109286e-05,
1148
+ "loss": 1.6736,
1149
+ "step": 1900
1150
+ },
1151
+ {
1152
+ "epoch": 1.43,
1153
+ "learning_rate": 2.622603933283545e-05,
1154
+ "loss": 1.6734,
1155
+ "step": 1910
1156
+ },
1157
+ {
1158
+ "epoch": 1.43,
1159
+ "learning_rate": 2.6101568334578048e-05,
1160
+ "loss": 1.6761,
1161
+ "step": 1920
1162
+ },
1163
+ {
1164
+ "epoch": 1.44,
1165
+ "learning_rate": 2.597709733632064e-05,
1166
+ "loss": 1.6842,
1167
+ "step": 1930
1168
+ },
1169
+ {
1170
+ "epoch": 1.45,
1171
+ "learning_rate": 2.5852626338063234e-05,
1172
+ "loss": 1.7064,
1173
+ "step": 1940
1174
+ },
1175
+ {
1176
+ "epoch": 1.46,
1177
+ "learning_rate": 2.5728155339805826e-05,
1178
+ "loss": 1.7036,
1179
+ "step": 1950
1180
+ },
1181
+ {
1182
+ "epoch": 1.46,
1183
+ "learning_rate": 2.5603684341548424e-05,
1184
+ "loss": 1.6563,
1185
+ "step": 1960
1186
+ },
1187
+ {
1188
+ "epoch": 1.47,
1189
+ "learning_rate": 2.5479213343291016e-05,
1190
+ "loss": 1.6796,
1191
+ "step": 1970
1192
+ },
1193
+ {
1194
+ "epoch": 1.48,
1195
+ "learning_rate": 2.535474234503361e-05,
1196
+ "loss": 1.6469,
1197
+ "step": 1980
1198
+ },
1199
+ {
1200
+ "epoch": 1.49,
1201
+ "learning_rate": 2.5230271346776202e-05,
1202
+ "loss": 1.6681,
1203
+ "step": 1990
1204
+ },
1205
+ {
1206
+ "epoch": 1.49,
1207
+ "learning_rate": 2.51058003485188e-05,
1208
+ "loss": 1.6632,
1209
+ "step": 2000
1210
+ },
1211
+ {
1212
+ "epoch": 1.5,
1213
+ "learning_rate": 2.498132935026139e-05,
1214
+ "loss": 1.6799,
1215
+ "step": 2010
1216
+ },
1217
+ {
1218
+ "epoch": 1.51,
1219
+ "learning_rate": 2.4856858352003983e-05,
1220
+ "loss": 1.6811,
1221
+ "step": 2020
1222
+ },
1223
+ {
1224
+ "epoch": 1.52,
1225
+ "learning_rate": 2.4732387353746578e-05,
1226
+ "loss": 1.6365,
1227
+ "step": 2030
1228
+ },
1229
+ {
1230
+ "epoch": 1.52,
1231
+ "learning_rate": 2.4607916355489173e-05,
1232
+ "loss": 1.6773,
1233
+ "step": 2040
1234
+ },
1235
+ {
1236
+ "epoch": 1.53,
1237
+ "learning_rate": 2.4483445357231764e-05,
1238
+ "loss": 1.6194,
1239
+ "step": 2050
1240
+ },
1241
+ {
1242
+ "epoch": 1.54,
1243
+ "learning_rate": 2.435897435897436e-05,
1244
+ "loss": 1.6865,
1245
+ "step": 2060
1246
+ },
1247
+ {
1248
+ "epoch": 1.55,
1249
+ "learning_rate": 2.4234503360716954e-05,
1250
+ "loss": 1.6552,
1251
+ "step": 2070
1252
+ },
1253
+ {
1254
+ "epoch": 1.55,
1255
+ "learning_rate": 2.411003236245955e-05,
1256
+ "loss": 1.6823,
1257
+ "step": 2080
1258
+ },
1259
+ {
1260
+ "epoch": 1.56,
1261
+ "learning_rate": 2.398556136420214e-05,
1262
+ "loss": 1.6611,
1263
+ "step": 2090
1264
+ },
1265
+ {
1266
+ "epoch": 1.57,
1267
+ "learning_rate": 2.3861090365944735e-05,
1268
+ "loss": 1.6662,
1269
+ "step": 2100
1270
+ },
1271
+ {
1272
+ "epoch": 1.58,
1273
+ "learning_rate": 2.373661936768733e-05,
1274
+ "loss": 1.6898,
1275
+ "step": 2110
1276
+ },
1277
+ {
1278
+ "epoch": 1.58,
1279
+ "learning_rate": 2.3612148369429922e-05,
1280
+ "loss": 1.6474,
1281
+ "step": 2120
1282
+ },
1283
+ {
1284
+ "epoch": 1.59,
1285
+ "learning_rate": 2.3487677371172517e-05,
1286
+ "loss": 1.6635,
1287
+ "step": 2130
1288
+ },
1289
+ {
1290
+ "epoch": 1.6,
1291
+ "learning_rate": 2.336320637291511e-05,
1292
+ "loss": 1.6274,
1293
+ "step": 2140
1294
+ },
1295
+ {
1296
+ "epoch": 1.61,
1297
+ "learning_rate": 2.3238735374657706e-05,
1298
+ "loss": 1.6794,
1299
+ "step": 2150
1300
+ },
1301
+ {
1302
+ "epoch": 1.61,
1303
+ "learning_rate": 2.3114264376400298e-05,
1304
+ "loss": 1.6411,
1305
+ "step": 2160
1306
+ },
1307
+ {
1308
+ "epoch": 1.62,
1309
+ "learning_rate": 2.2989793378142893e-05,
1310
+ "loss": 1.6793,
1311
+ "step": 2170
1312
+ },
1313
+ {
1314
+ "epoch": 1.63,
1315
+ "learning_rate": 2.2865322379885488e-05,
1316
+ "loss": 1.6772,
1317
+ "step": 2180
1318
+ },
1319
+ {
1320
+ "epoch": 1.64,
1321
+ "learning_rate": 2.2740851381628083e-05,
1322
+ "loss": 1.6425,
1323
+ "step": 2190
1324
+ },
1325
+ {
1326
+ "epoch": 1.64,
1327
+ "learning_rate": 2.2616380383370674e-05,
1328
+ "loss": 1.6614,
1329
+ "step": 2200
1330
+ },
1331
+ {
1332
+ "epoch": 1.65,
1333
+ "learning_rate": 2.249190938511327e-05,
1334
+ "loss": 1.65,
1335
+ "step": 2210
1336
+ },
1337
+ {
1338
+ "epoch": 1.66,
1339
+ "learning_rate": 2.2367438386855864e-05,
1340
+ "loss": 1.685,
1341
+ "step": 2220
1342
+ },
1343
+ {
1344
+ "epoch": 1.67,
1345
+ "learning_rate": 2.2242967388598455e-05,
1346
+ "loss": 1.6808,
1347
+ "step": 2230
1348
+ },
1349
+ {
1350
+ "epoch": 1.67,
1351
+ "learning_rate": 2.211849639034105e-05,
1352
+ "loss": 1.6612,
1353
+ "step": 2240
1354
+ },
1355
+ {
1356
+ "epoch": 1.68,
1357
+ "learning_rate": 2.1994025392083645e-05,
1358
+ "loss": 1.6631,
1359
+ "step": 2250
1360
+ },
1361
+ {
1362
+ "epoch": 1.69,
1363
+ "learning_rate": 2.186955439382624e-05,
1364
+ "loss": 1.6675,
1365
+ "step": 2260
1366
+ },
1367
+ {
1368
+ "epoch": 1.7,
1369
+ "learning_rate": 2.174508339556883e-05,
1370
+ "loss": 1.6532,
1371
+ "step": 2270
1372
+ },
1373
+ {
1374
+ "epoch": 1.7,
1375
+ "learning_rate": 2.1620612397311426e-05,
1376
+ "loss": 1.7217,
1377
+ "step": 2280
1378
+ },
1379
+ {
1380
+ "epoch": 1.71,
1381
+ "learning_rate": 2.149614139905402e-05,
1382
+ "loss": 1.6697,
1383
+ "step": 2290
1384
+ },
1385
+ {
1386
+ "epoch": 1.72,
1387
+ "learning_rate": 2.1371670400796616e-05,
1388
+ "loss": 1.634,
1389
+ "step": 2300
1390
+ },
1391
+ {
1392
+ "epoch": 1.73,
1393
+ "learning_rate": 2.1247199402539207e-05,
1394
+ "loss": 1.6968,
1395
+ "step": 2310
1396
+ },
1397
+ {
1398
+ "epoch": 1.73,
1399
+ "learning_rate": 2.1122728404281802e-05,
1400
+ "loss": 1.6925,
1401
+ "step": 2320
1402
+ },
1403
+ {
1404
+ "epoch": 1.74,
1405
+ "learning_rate": 2.0998257406024397e-05,
1406
+ "loss": 1.6455,
1407
+ "step": 2330
1408
+ },
1409
+ {
1410
+ "epoch": 1.75,
1411
+ "learning_rate": 2.0873786407766992e-05,
1412
+ "loss": 1.6365,
1413
+ "step": 2340
1414
+ },
1415
+ {
1416
+ "epoch": 1.76,
1417
+ "learning_rate": 2.0749315409509583e-05,
1418
+ "loss": 1.6813,
1419
+ "step": 2350
1420
+ },
1421
+ {
1422
+ "epoch": 1.76,
1423
+ "learning_rate": 2.062484441125218e-05,
1424
+ "loss": 1.6725,
1425
+ "step": 2360
1426
+ },
1427
+ {
1428
+ "epoch": 1.77,
1429
+ "learning_rate": 2.0500373412994773e-05,
1430
+ "loss": 1.6718,
1431
+ "step": 2370
1432
+ },
1433
+ {
1434
+ "epoch": 1.78,
1435
+ "learning_rate": 2.0375902414737365e-05,
1436
+ "loss": 1.6954,
1437
+ "step": 2380
1438
+ },
1439
+ {
1440
+ "epoch": 1.78,
1441
+ "learning_rate": 2.025143141647996e-05,
1442
+ "loss": 1.6719,
1443
+ "step": 2390
1444
+ },
1445
+ {
1446
+ "epoch": 1.79,
1447
+ "learning_rate": 2.0126960418222554e-05,
1448
+ "loss": 1.6549,
1449
+ "step": 2400
1450
+ },
1451
+ {
1452
+ "epoch": 1.8,
1453
+ "learning_rate": 2.000248941996515e-05,
1454
+ "loss": 1.6761,
1455
+ "step": 2410
1456
+ },
1457
+ {
1458
+ "epoch": 1.81,
1459
+ "learning_rate": 1.987801842170774e-05,
1460
+ "loss": 1.6582,
1461
+ "step": 2420
1462
+ },
1463
+ {
1464
+ "epoch": 1.81,
1465
+ "learning_rate": 1.9753547423450336e-05,
1466
+ "loss": 1.6202,
1467
+ "step": 2430
1468
+ },
1469
+ {
1470
+ "epoch": 1.82,
1471
+ "learning_rate": 1.962907642519293e-05,
1472
+ "loss": 1.6645,
1473
+ "step": 2440
1474
+ },
1475
+ {
1476
+ "epoch": 1.83,
1477
+ "learning_rate": 1.9504605426935525e-05,
1478
+ "loss": 1.6565,
1479
+ "step": 2450
1480
+ },
1481
+ {
1482
+ "epoch": 1.84,
1483
+ "learning_rate": 1.9380134428678117e-05,
1484
+ "loss": 1.6395,
1485
+ "step": 2460
1486
+ },
1487
+ {
1488
+ "epoch": 1.84,
1489
+ "learning_rate": 1.9255663430420712e-05,
1490
+ "loss": 1.6469,
1491
+ "step": 2470
1492
+ },
1493
+ {
1494
+ "epoch": 1.85,
1495
+ "learning_rate": 1.9131192432163307e-05,
1496
+ "loss": 1.6523,
1497
+ "step": 2480
1498
+ },
1499
+ {
1500
+ "epoch": 1.86,
1501
+ "learning_rate": 1.9006721433905898e-05,
1502
+ "loss": 1.6799,
1503
+ "step": 2490
1504
+ },
1505
+ {
1506
+ "epoch": 1.87,
1507
+ "learning_rate": 1.8882250435648493e-05,
1508
+ "loss": 1.6584,
1509
+ "step": 2500
1510
+ }
1511
+ ],
1512
+ "logging_steps": 10,
1513
+ "max_steps": 4017,
1514
+ "num_input_tokens_seen": 0,
1515
+ "num_train_epochs": 3,
1516
+ "save_steps": 500,
1517
+ "total_flos": 1.5800620548096e+18,
1518
+ "train_batch_size": 4,
1519
+ "trial_name": null,
1520
+ "trial_params": null
1521
+ }
119_Llama-2-13b-hf/checkpoint-2500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:95fdd19e2ed6f33177f737bfa3ccdb7380f84c750beb151600638af981e7df02
3
+ size 4283
119_Llama-2-13b-hf/checkpoint-3000/README.md ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+ ### Framework versions
7
+
8
+
9
+ - PEFT 0.4.0
119_Llama-2-13b-hf/checkpoint-3000/adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "meta-llama/Llama-2-13b-hf",
4
+ "bias": "none",
5
+ "fan_in_fan_out": false,
6
+ "inference_mode": true,
7
+ "init_lora_weights": true,
8
+ "layers_pattern": null,
9
+ "layers_to_transform": null,
10
+ "lora_alpha": 32,
11
+ "lora_dropout": 0.05,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 8,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "q_proj",
18
+ "v_proj"
19
+ ],
20
+ "task_type": "CAUSAL_LM"
21
+ }
119_Llama-2-13b-hf/checkpoint-3000/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0aa1cf1d60b91c4e580817946538f8310928082df42efac7d9bf7a3110c92395
3
+ size 26235704
119_Llama-2-13b-hf/checkpoint-3000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:da1dd1ff8fbfa1c6771379adfff130d77c3fa48af0811c4e58131eb2b9b3a145
3
+ size 52567237
119_Llama-2-13b-hf/checkpoint-3000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:767078b8ef1ffe7f2a6277af0f9d2f9ce60b449625e7e1001933c38f3d9b5044
3
+ size 14575
119_Llama-2-13b-hf/checkpoint-3000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3ccf9abe8cc59bf312498a95618303458af5f7d39db38f6bfa2ade359720f934
3
+ size 627
119_Llama-2-13b-hf/checkpoint-3000/trainer_state.json ADDED
@@ -0,0 +1,1821 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.2404779686333085,
5
+ "eval_steps": 500,
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": 0.01,
13
+ "learning_rate": 4.98755290017426e-05,
14
+ "loss": 1.7891,
15
+ "step": 10
16
+ },
17
+ {
18
+ "epoch": 0.01,
19
+ "learning_rate": 4.975105800348519e-05,
20
+ "loss": 1.7869,
21
+ "step": 20
22
+ },
23
+ {
24
+ "epoch": 0.02,
25
+ "learning_rate": 4.962658700522779e-05,
26
+ "loss": 1.7868,
27
+ "step": 30
28
+ },
29
+ {
30
+ "epoch": 0.03,
31
+ "learning_rate": 4.9502116006970375e-05,
32
+ "loss": 1.7547,
33
+ "step": 40
34
+ },
35
+ {
36
+ "epoch": 0.04,
37
+ "learning_rate": 4.937764500871298e-05,
38
+ "loss": 1.6772,
39
+ "step": 50
40
+ },
41
+ {
42
+ "epoch": 0.04,
43
+ "learning_rate": 4.9253174010455565e-05,
44
+ "loss": 1.7076,
45
+ "step": 60
46
+ },
47
+ {
48
+ "epoch": 0.05,
49
+ "learning_rate": 4.912870301219816e-05,
50
+ "loss": 1.6948,
51
+ "step": 70
52
+ },
53
+ {
54
+ "epoch": 0.06,
55
+ "learning_rate": 4.9004232013940755e-05,
56
+ "loss": 1.7135,
57
+ "step": 80
58
+ },
59
+ {
60
+ "epoch": 0.07,
61
+ "learning_rate": 4.887976101568335e-05,
62
+ "loss": 1.7007,
63
+ "step": 90
64
+ },
65
+ {
66
+ "epoch": 0.07,
67
+ "learning_rate": 4.8755290017425944e-05,
68
+ "loss": 1.7247,
69
+ "step": 100
70
+ },
71
+ {
72
+ "epoch": 0.08,
73
+ "learning_rate": 4.863081901916854e-05,
74
+ "loss": 1.719,
75
+ "step": 110
76
+ },
77
+ {
78
+ "epoch": 0.09,
79
+ "learning_rate": 4.850634802091113e-05,
80
+ "loss": 1.7209,
81
+ "step": 120
82
+ },
83
+ {
84
+ "epoch": 0.1,
85
+ "learning_rate": 4.838187702265373e-05,
86
+ "loss": 1.7354,
87
+ "step": 130
88
+ },
89
+ {
90
+ "epoch": 0.1,
91
+ "learning_rate": 4.825740602439632e-05,
92
+ "loss": 1.7025,
93
+ "step": 140
94
+ },
95
+ {
96
+ "epoch": 0.11,
97
+ "learning_rate": 4.813293502613891e-05,
98
+ "loss": 1.6664,
99
+ "step": 150
100
+ },
101
+ {
102
+ "epoch": 0.12,
103
+ "learning_rate": 4.800846402788151e-05,
104
+ "loss": 1.6851,
105
+ "step": 160
106
+ },
107
+ {
108
+ "epoch": 0.13,
109
+ "learning_rate": 4.78839930296241e-05,
110
+ "loss": 1.6749,
111
+ "step": 170
112
+ },
113
+ {
114
+ "epoch": 0.13,
115
+ "learning_rate": 4.7759522031366697e-05,
116
+ "loss": 1.6887,
117
+ "step": 180
118
+ },
119
+ {
120
+ "epoch": 0.14,
121
+ "learning_rate": 4.763505103310929e-05,
122
+ "loss": 1.6571,
123
+ "step": 190
124
+ },
125
+ {
126
+ "epoch": 0.15,
127
+ "learning_rate": 4.751058003485188e-05,
128
+ "loss": 1.7136,
129
+ "step": 200
130
+ },
131
+ {
132
+ "epoch": 0.16,
133
+ "learning_rate": 4.738610903659448e-05,
134
+ "loss": 1.7098,
135
+ "step": 210
136
+ },
137
+ {
138
+ "epoch": 0.16,
139
+ "learning_rate": 4.726163803833707e-05,
140
+ "loss": 1.7285,
141
+ "step": 220
142
+ },
143
+ {
144
+ "epoch": 0.17,
145
+ "learning_rate": 4.7137167040079664e-05,
146
+ "loss": 1.7038,
147
+ "step": 230
148
+ },
149
+ {
150
+ "epoch": 0.18,
151
+ "learning_rate": 4.701269604182226e-05,
152
+ "loss": 1.692,
153
+ "step": 240
154
+ },
155
+ {
156
+ "epoch": 0.19,
157
+ "learning_rate": 4.6888225043564854e-05,
158
+ "loss": 1.6825,
159
+ "step": 250
160
+ },
161
+ {
162
+ "epoch": 0.19,
163
+ "learning_rate": 4.676375404530744e-05,
164
+ "loss": 1.7254,
165
+ "step": 260
166
+ },
167
+ {
168
+ "epoch": 0.2,
169
+ "learning_rate": 4.6639283047050044e-05,
170
+ "loss": 1.6765,
171
+ "step": 270
172
+ },
173
+ {
174
+ "epoch": 0.21,
175
+ "learning_rate": 4.651481204879263e-05,
176
+ "loss": 1.6796,
177
+ "step": 280
178
+ },
179
+ {
180
+ "epoch": 0.22,
181
+ "learning_rate": 4.6390341050535227e-05,
182
+ "loss": 1.7302,
183
+ "step": 290
184
+ },
185
+ {
186
+ "epoch": 0.22,
187
+ "learning_rate": 4.626587005227782e-05,
188
+ "loss": 1.7469,
189
+ "step": 300
190
+ },
191
+ {
192
+ "epoch": 0.23,
193
+ "learning_rate": 4.6141399054020416e-05,
194
+ "loss": 1.6345,
195
+ "step": 310
196
+ },
197
+ {
198
+ "epoch": 0.24,
199
+ "learning_rate": 4.601692805576301e-05,
200
+ "loss": 1.6761,
201
+ "step": 320
202
+ },
203
+ {
204
+ "epoch": 0.25,
205
+ "learning_rate": 4.5892457057505606e-05,
206
+ "loss": 1.6602,
207
+ "step": 330
208
+ },
209
+ {
210
+ "epoch": 0.25,
211
+ "learning_rate": 4.5767986059248194e-05,
212
+ "loss": 1.6637,
213
+ "step": 340
214
+ },
215
+ {
216
+ "epoch": 0.26,
217
+ "learning_rate": 4.5643515060990796e-05,
218
+ "loss": 1.6953,
219
+ "step": 350
220
+ },
221
+ {
222
+ "epoch": 0.27,
223
+ "learning_rate": 4.5519044062733384e-05,
224
+ "loss": 1.6829,
225
+ "step": 360
226
+ },
227
+ {
228
+ "epoch": 0.28,
229
+ "learning_rate": 4.539457306447598e-05,
230
+ "loss": 1.657,
231
+ "step": 370
232
+ },
233
+ {
234
+ "epoch": 0.28,
235
+ "learning_rate": 4.5270102066218574e-05,
236
+ "loss": 1.647,
237
+ "step": 380
238
+ },
239
+ {
240
+ "epoch": 0.29,
241
+ "learning_rate": 4.514563106796117e-05,
242
+ "loss": 1.6854,
243
+ "step": 390
244
+ },
245
+ {
246
+ "epoch": 0.3,
247
+ "learning_rate": 4.502116006970376e-05,
248
+ "loss": 1.6737,
249
+ "step": 400
250
+ },
251
+ {
252
+ "epoch": 0.31,
253
+ "learning_rate": 4.489668907144636e-05,
254
+ "loss": 1.6938,
255
+ "step": 410
256
+ },
257
+ {
258
+ "epoch": 0.31,
259
+ "learning_rate": 4.4772218073188946e-05,
260
+ "loss": 1.6721,
261
+ "step": 420
262
+ },
263
+ {
264
+ "epoch": 0.32,
265
+ "learning_rate": 4.464774707493155e-05,
266
+ "loss": 1.6683,
267
+ "step": 430
268
+ },
269
+ {
270
+ "epoch": 0.33,
271
+ "learning_rate": 4.4523276076674136e-05,
272
+ "loss": 1.6229,
273
+ "step": 440
274
+ },
275
+ {
276
+ "epoch": 0.34,
277
+ "learning_rate": 4.439880507841673e-05,
278
+ "loss": 1.6593,
279
+ "step": 450
280
+ },
281
+ {
282
+ "epoch": 0.34,
283
+ "learning_rate": 4.4274334080159326e-05,
284
+ "loss": 1.6844,
285
+ "step": 460
286
+ },
287
+ {
288
+ "epoch": 0.35,
289
+ "learning_rate": 4.414986308190192e-05,
290
+ "loss": 1.6754,
291
+ "step": 470
292
+ },
293
+ {
294
+ "epoch": 0.36,
295
+ "learning_rate": 4.4025392083644516e-05,
296
+ "loss": 1.6747,
297
+ "step": 480
298
+ },
299
+ {
300
+ "epoch": 0.37,
301
+ "learning_rate": 4.390092108538711e-05,
302
+ "loss": 1.6789,
303
+ "step": 490
304
+ },
305
+ {
306
+ "epoch": 0.37,
307
+ "learning_rate": 4.37764500871297e-05,
308
+ "loss": 1.6956,
309
+ "step": 500
310
+ },
311
+ {
312
+ "epoch": 0.38,
313
+ "learning_rate": 4.36519790888723e-05,
314
+ "loss": 1.689,
315
+ "step": 510
316
+ },
317
+ {
318
+ "epoch": 0.39,
319
+ "learning_rate": 4.352750809061489e-05,
320
+ "loss": 1.673,
321
+ "step": 520
322
+ },
323
+ {
324
+ "epoch": 0.4,
325
+ "learning_rate": 4.340303709235748e-05,
326
+ "loss": 1.6968,
327
+ "step": 530
328
+ },
329
+ {
330
+ "epoch": 0.4,
331
+ "learning_rate": 4.327856609410008e-05,
332
+ "loss": 1.6569,
333
+ "step": 540
334
+ },
335
+ {
336
+ "epoch": 0.41,
337
+ "learning_rate": 4.315409509584267e-05,
338
+ "loss": 1.6777,
339
+ "step": 550
340
+ },
341
+ {
342
+ "epoch": 0.42,
343
+ "learning_rate": 4.302962409758526e-05,
344
+ "loss": 1.71,
345
+ "step": 560
346
+ },
347
+ {
348
+ "epoch": 0.43,
349
+ "learning_rate": 4.290515309932786e-05,
350
+ "loss": 1.6943,
351
+ "step": 570
352
+ },
353
+ {
354
+ "epoch": 0.43,
355
+ "learning_rate": 4.278068210107045e-05,
356
+ "loss": 1.6925,
357
+ "step": 580
358
+ },
359
+ {
360
+ "epoch": 0.44,
361
+ "learning_rate": 4.2656211102813046e-05,
362
+ "loss": 1.6628,
363
+ "step": 590
364
+ },
365
+ {
366
+ "epoch": 0.45,
367
+ "learning_rate": 4.253174010455564e-05,
368
+ "loss": 1.7123,
369
+ "step": 600
370
+ },
371
+ {
372
+ "epoch": 0.46,
373
+ "learning_rate": 4.2407269106298235e-05,
374
+ "loss": 1.6772,
375
+ "step": 610
376
+ },
377
+ {
378
+ "epoch": 0.46,
379
+ "learning_rate": 4.228279810804083e-05,
380
+ "loss": 1.6429,
381
+ "step": 620
382
+ },
383
+ {
384
+ "epoch": 0.47,
385
+ "learning_rate": 4.2158327109783425e-05,
386
+ "loss": 1.6472,
387
+ "step": 630
388
+ },
389
+ {
390
+ "epoch": 0.48,
391
+ "learning_rate": 4.203385611152601e-05,
392
+ "loss": 1.6825,
393
+ "step": 640
394
+ },
395
+ {
396
+ "epoch": 0.49,
397
+ "learning_rate": 4.1909385113268615e-05,
398
+ "loss": 1.6825,
399
+ "step": 650
400
+ },
401
+ {
402
+ "epoch": 0.49,
403
+ "learning_rate": 4.17849141150112e-05,
404
+ "loss": 1.6818,
405
+ "step": 660
406
+ },
407
+ {
408
+ "epoch": 0.5,
409
+ "learning_rate": 4.16604431167538e-05,
410
+ "loss": 1.7325,
411
+ "step": 670
412
+ },
413
+ {
414
+ "epoch": 0.51,
415
+ "learning_rate": 4.153597211849639e-05,
416
+ "loss": 1.7008,
417
+ "step": 680
418
+ },
419
+ {
420
+ "epoch": 0.52,
421
+ "learning_rate": 4.141150112023899e-05,
422
+ "loss": 1.655,
423
+ "step": 690
424
+ },
425
+ {
426
+ "epoch": 0.52,
427
+ "learning_rate": 4.128703012198158e-05,
428
+ "loss": 1.6718,
429
+ "step": 700
430
+ },
431
+ {
432
+ "epoch": 0.53,
433
+ "learning_rate": 4.116255912372418e-05,
434
+ "loss": 1.6412,
435
+ "step": 710
436
+ },
437
+ {
438
+ "epoch": 0.54,
439
+ "learning_rate": 4.1038088125466765e-05,
440
+ "loss": 1.7091,
441
+ "step": 720
442
+ },
443
+ {
444
+ "epoch": 0.55,
445
+ "learning_rate": 4.091361712720937e-05,
446
+ "loss": 1.726,
447
+ "step": 730
448
+ },
449
+ {
450
+ "epoch": 0.55,
451
+ "learning_rate": 4.0789146128951955e-05,
452
+ "loss": 1.6655,
453
+ "step": 740
454
+ },
455
+ {
456
+ "epoch": 0.56,
457
+ "learning_rate": 4.066467513069455e-05,
458
+ "loss": 1.7242,
459
+ "step": 750
460
+ },
461
+ {
462
+ "epoch": 0.57,
463
+ "learning_rate": 4.0540204132437145e-05,
464
+ "loss": 1.6714,
465
+ "step": 760
466
+ },
467
+ {
468
+ "epoch": 0.58,
469
+ "learning_rate": 4.041573313417974e-05,
470
+ "loss": 1.6646,
471
+ "step": 770
472
+ },
473
+ {
474
+ "epoch": 0.58,
475
+ "learning_rate": 4.029126213592233e-05,
476
+ "loss": 1.6809,
477
+ "step": 780
478
+ },
479
+ {
480
+ "epoch": 0.59,
481
+ "learning_rate": 4.016679113766493e-05,
482
+ "loss": 1.6868,
483
+ "step": 790
484
+ },
485
+ {
486
+ "epoch": 0.6,
487
+ "learning_rate": 4.004232013940752e-05,
488
+ "loss": 1.694,
489
+ "step": 800
490
+ },
491
+ {
492
+ "epoch": 0.6,
493
+ "learning_rate": 3.991784914115011e-05,
494
+ "loss": 1.6801,
495
+ "step": 810
496
+ },
497
+ {
498
+ "epoch": 0.61,
499
+ "learning_rate": 3.979337814289271e-05,
500
+ "loss": 1.6861,
501
+ "step": 820
502
+ },
503
+ {
504
+ "epoch": 0.62,
505
+ "learning_rate": 3.96689071446353e-05,
506
+ "loss": 1.6646,
507
+ "step": 830
508
+ },
509
+ {
510
+ "epoch": 0.63,
511
+ "learning_rate": 3.95444361463779e-05,
512
+ "loss": 1.6764,
513
+ "step": 840
514
+ },
515
+ {
516
+ "epoch": 0.63,
517
+ "learning_rate": 3.941996514812049e-05,
518
+ "loss": 1.6821,
519
+ "step": 850
520
+ },
521
+ {
522
+ "epoch": 0.64,
523
+ "learning_rate": 3.929549414986308e-05,
524
+ "loss": 1.6681,
525
+ "step": 860
526
+ },
527
+ {
528
+ "epoch": 0.65,
529
+ "learning_rate": 3.917102315160568e-05,
530
+ "loss": 1.6847,
531
+ "step": 870
532
+ },
533
+ {
534
+ "epoch": 0.66,
535
+ "learning_rate": 3.904655215334827e-05,
536
+ "loss": 1.6084,
537
+ "step": 880
538
+ },
539
+ {
540
+ "epoch": 0.66,
541
+ "learning_rate": 3.8922081155090865e-05,
542
+ "loss": 1.6624,
543
+ "step": 890
544
+ },
545
+ {
546
+ "epoch": 0.67,
547
+ "learning_rate": 3.879761015683346e-05,
548
+ "loss": 1.6619,
549
+ "step": 900
550
+ },
551
+ {
552
+ "epoch": 0.68,
553
+ "learning_rate": 3.8673139158576054e-05,
554
+ "loss": 1.7024,
555
+ "step": 910
556
+ },
557
+ {
558
+ "epoch": 0.69,
559
+ "learning_rate": 3.854866816031865e-05,
560
+ "loss": 1.7301,
561
+ "step": 920
562
+ },
563
+ {
564
+ "epoch": 0.69,
565
+ "learning_rate": 3.8424197162061244e-05,
566
+ "loss": 1.6907,
567
+ "step": 930
568
+ },
569
+ {
570
+ "epoch": 0.7,
571
+ "learning_rate": 3.829972616380383e-05,
572
+ "loss": 1.7046,
573
+ "step": 940
574
+ },
575
+ {
576
+ "epoch": 0.71,
577
+ "learning_rate": 3.8175255165546434e-05,
578
+ "loss": 1.6661,
579
+ "step": 950
580
+ },
581
+ {
582
+ "epoch": 0.72,
583
+ "learning_rate": 3.805078416728902e-05,
584
+ "loss": 1.6981,
585
+ "step": 960
586
+ },
587
+ {
588
+ "epoch": 0.72,
589
+ "learning_rate": 3.792631316903162e-05,
590
+ "loss": 1.662,
591
+ "step": 970
592
+ },
593
+ {
594
+ "epoch": 0.73,
595
+ "learning_rate": 3.780184217077421e-05,
596
+ "loss": 1.7073,
597
+ "step": 980
598
+ },
599
+ {
600
+ "epoch": 0.74,
601
+ "learning_rate": 3.7677371172516806e-05,
602
+ "loss": 1.6488,
603
+ "step": 990
604
+ },
605
+ {
606
+ "epoch": 0.75,
607
+ "learning_rate": 3.7552900174259395e-05,
608
+ "loss": 1.6708,
609
+ "step": 1000
610
+ },
611
+ {
612
+ "epoch": 0.75,
613
+ "learning_rate": 3.7428429176001996e-05,
614
+ "loss": 1.7045,
615
+ "step": 1010
616
+ },
617
+ {
618
+ "epoch": 0.76,
619
+ "learning_rate": 3.7303958177744584e-05,
620
+ "loss": 1.6842,
621
+ "step": 1020
622
+ },
623
+ {
624
+ "epoch": 0.77,
625
+ "learning_rate": 3.717948717948718e-05,
626
+ "loss": 1.6682,
627
+ "step": 1030
628
+ },
629
+ {
630
+ "epoch": 0.78,
631
+ "learning_rate": 3.7055016181229774e-05,
632
+ "loss": 1.636,
633
+ "step": 1040
634
+ },
635
+ {
636
+ "epoch": 0.78,
637
+ "learning_rate": 3.693054518297237e-05,
638
+ "loss": 1.6801,
639
+ "step": 1050
640
+ },
641
+ {
642
+ "epoch": 0.79,
643
+ "learning_rate": 3.6806074184714964e-05,
644
+ "loss": 1.6371,
645
+ "step": 1060
646
+ },
647
+ {
648
+ "epoch": 0.8,
649
+ "learning_rate": 3.668160318645756e-05,
650
+ "loss": 1.6755,
651
+ "step": 1070
652
+ },
653
+ {
654
+ "epoch": 0.81,
655
+ "learning_rate": 3.655713218820015e-05,
656
+ "loss": 1.7046,
657
+ "step": 1080
658
+ },
659
+ {
660
+ "epoch": 0.81,
661
+ "learning_rate": 3.643266118994275e-05,
662
+ "loss": 1.707,
663
+ "step": 1090
664
+ },
665
+ {
666
+ "epoch": 0.82,
667
+ "learning_rate": 3.6308190191685337e-05,
668
+ "loss": 1.6568,
669
+ "step": 1100
670
+ },
671
+ {
672
+ "epoch": 0.83,
673
+ "learning_rate": 3.618371919342793e-05,
674
+ "loss": 1.6447,
675
+ "step": 1110
676
+ },
677
+ {
678
+ "epoch": 0.84,
679
+ "learning_rate": 3.6059248195170526e-05,
680
+ "loss": 1.691,
681
+ "step": 1120
682
+ },
683
+ {
684
+ "epoch": 0.84,
685
+ "learning_rate": 3.593477719691312e-05,
686
+ "loss": 1.6747,
687
+ "step": 1130
688
+ },
689
+ {
690
+ "epoch": 0.85,
691
+ "learning_rate": 3.5810306198655716e-05,
692
+ "loss": 1.656,
693
+ "step": 1140
694
+ },
695
+ {
696
+ "epoch": 0.86,
697
+ "learning_rate": 3.568583520039831e-05,
698
+ "loss": 1.6485,
699
+ "step": 1150
700
+ },
701
+ {
702
+ "epoch": 0.87,
703
+ "learning_rate": 3.55613642021409e-05,
704
+ "loss": 1.642,
705
+ "step": 1160
706
+ },
707
+ {
708
+ "epoch": 0.87,
709
+ "learning_rate": 3.54368932038835e-05,
710
+ "loss": 1.6891,
711
+ "step": 1170
712
+ },
713
+ {
714
+ "epoch": 0.88,
715
+ "learning_rate": 3.531242220562609e-05,
716
+ "loss": 1.716,
717
+ "step": 1180
718
+ },
719
+ {
720
+ "epoch": 0.89,
721
+ "learning_rate": 3.5187951207368684e-05,
722
+ "loss": 1.6785,
723
+ "step": 1190
724
+ },
725
+ {
726
+ "epoch": 0.9,
727
+ "learning_rate": 3.506348020911128e-05,
728
+ "loss": 1.7053,
729
+ "step": 1200
730
+ },
731
+ {
732
+ "epoch": 0.9,
733
+ "learning_rate": 3.493900921085387e-05,
734
+ "loss": 1.6833,
735
+ "step": 1210
736
+ },
737
+ {
738
+ "epoch": 0.91,
739
+ "learning_rate": 3.481453821259646e-05,
740
+ "loss": 1.7336,
741
+ "step": 1220
742
+ },
743
+ {
744
+ "epoch": 0.92,
745
+ "learning_rate": 3.469006721433906e-05,
746
+ "loss": 1.6496,
747
+ "step": 1230
748
+ },
749
+ {
750
+ "epoch": 0.93,
751
+ "learning_rate": 3.456559621608165e-05,
752
+ "loss": 1.7014,
753
+ "step": 1240
754
+ },
755
+ {
756
+ "epoch": 0.93,
757
+ "learning_rate": 3.4441125217824246e-05,
758
+ "loss": 1.6502,
759
+ "step": 1250
760
+ },
761
+ {
762
+ "epoch": 0.94,
763
+ "learning_rate": 3.431665421956684e-05,
764
+ "loss": 1.6617,
765
+ "step": 1260
766
+ },
767
+ {
768
+ "epoch": 0.95,
769
+ "learning_rate": 3.4192183221309436e-05,
770
+ "loss": 1.7155,
771
+ "step": 1270
772
+ },
773
+ {
774
+ "epoch": 0.96,
775
+ "learning_rate": 3.406771222305203e-05,
776
+ "loss": 1.691,
777
+ "step": 1280
778
+ },
779
+ {
780
+ "epoch": 0.96,
781
+ "learning_rate": 3.3943241224794625e-05,
782
+ "loss": 1.6601,
783
+ "step": 1290
784
+ },
785
+ {
786
+ "epoch": 0.97,
787
+ "learning_rate": 3.3818770226537214e-05,
788
+ "loss": 1.6607,
789
+ "step": 1300
790
+ },
791
+ {
792
+ "epoch": 0.98,
793
+ "learning_rate": 3.3694299228279815e-05,
794
+ "loss": 1.7053,
795
+ "step": 1310
796
+ },
797
+ {
798
+ "epoch": 0.99,
799
+ "learning_rate": 3.35698282300224e-05,
800
+ "loss": 1.6681,
801
+ "step": 1320
802
+ },
803
+ {
804
+ "epoch": 0.99,
805
+ "learning_rate": 3.3445357231765e-05,
806
+ "loss": 1.6712,
807
+ "step": 1330
808
+ },
809
+ {
810
+ "epoch": 1.0,
811
+ "learning_rate": 3.332088623350759e-05,
812
+ "loss": 1.6601,
813
+ "step": 1340
814
+ },
815
+ {
816
+ "epoch": 1.01,
817
+ "learning_rate": 3.319641523525019e-05,
818
+ "loss": 1.6522,
819
+ "step": 1350
820
+ },
821
+ {
822
+ "epoch": 1.02,
823
+ "learning_rate": 3.307194423699278e-05,
824
+ "loss": 1.6621,
825
+ "step": 1360
826
+ },
827
+ {
828
+ "epoch": 1.02,
829
+ "learning_rate": 3.294747323873538e-05,
830
+ "loss": 1.6651,
831
+ "step": 1370
832
+ },
833
+ {
834
+ "epoch": 1.03,
835
+ "learning_rate": 3.2823002240477966e-05,
836
+ "loss": 1.6665,
837
+ "step": 1380
838
+ },
839
+ {
840
+ "epoch": 1.04,
841
+ "learning_rate": 3.269853124222057e-05,
842
+ "loss": 1.6811,
843
+ "step": 1390
844
+ },
845
+ {
846
+ "epoch": 1.05,
847
+ "learning_rate": 3.2574060243963156e-05,
848
+ "loss": 1.6632,
849
+ "step": 1400
850
+ },
851
+ {
852
+ "epoch": 1.05,
853
+ "learning_rate": 3.244958924570575e-05,
854
+ "loss": 1.6529,
855
+ "step": 1410
856
+ },
857
+ {
858
+ "epoch": 1.06,
859
+ "learning_rate": 3.2325118247448345e-05,
860
+ "loss": 1.6943,
861
+ "step": 1420
862
+ },
863
+ {
864
+ "epoch": 1.07,
865
+ "learning_rate": 3.220064724919094e-05,
866
+ "loss": 1.6932,
867
+ "step": 1430
868
+ },
869
+ {
870
+ "epoch": 1.08,
871
+ "learning_rate": 3.2076176250933535e-05,
872
+ "loss": 1.6711,
873
+ "step": 1440
874
+ },
875
+ {
876
+ "epoch": 1.08,
877
+ "learning_rate": 3.195170525267613e-05,
878
+ "loss": 1.6711,
879
+ "step": 1450
880
+ },
881
+ {
882
+ "epoch": 1.09,
883
+ "learning_rate": 3.182723425441872e-05,
884
+ "loss": 1.6678,
885
+ "step": 1460
886
+ },
887
+ {
888
+ "epoch": 1.1,
889
+ "learning_rate": 3.170276325616132e-05,
890
+ "loss": 1.7058,
891
+ "step": 1470
892
+ },
893
+ {
894
+ "epoch": 1.11,
895
+ "learning_rate": 3.157829225790391e-05,
896
+ "loss": 1.7017,
897
+ "step": 1480
898
+ },
899
+ {
900
+ "epoch": 1.11,
901
+ "learning_rate": 3.14538212596465e-05,
902
+ "loss": 1.6862,
903
+ "step": 1490
904
+ },
905
+ {
906
+ "epoch": 1.12,
907
+ "learning_rate": 3.13293502613891e-05,
908
+ "loss": 1.6425,
909
+ "step": 1500
910
+ },
911
+ {
912
+ "epoch": 1.13,
913
+ "learning_rate": 3.120487926313169e-05,
914
+ "loss": 1.6977,
915
+ "step": 1510
916
+ },
917
+ {
918
+ "epoch": 1.14,
919
+ "learning_rate": 3.108040826487428e-05,
920
+ "loss": 1.698,
921
+ "step": 1520
922
+ },
923
+ {
924
+ "epoch": 1.14,
925
+ "learning_rate": 3.095593726661688e-05,
926
+ "loss": 1.6718,
927
+ "step": 1530
928
+ },
929
+ {
930
+ "epoch": 1.15,
931
+ "learning_rate": 3.083146626835947e-05,
932
+ "loss": 1.7025,
933
+ "step": 1540
934
+ },
935
+ {
936
+ "epoch": 1.16,
937
+ "learning_rate": 3.0706995270102065e-05,
938
+ "loss": 1.665,
939
+ "step": 1550
940
+ },
941
+ {
942
+ "epoch": 1.17,
943
+ "learning_rate": 3.058252427184466e-05,
944
+ "loss": 1.7035,
945
+ "step": 1560
946
+ },
947
+ {
948
+ "epoch": 1.17,
949
+ "learning_rate": 3.0458053273587255e-05,
950
+ "loss": 1.6849,
951
+ "step": 1570
952
+ },
953
+ {
954
+ "epoch": 1.18,
955
+ "learning_rate": 3.0333582275329846e-05,
956
+ "loss": 1.6345,
957
+ "step": 1580
958
+ },
959
+ {
960
+ "epoch": 1.19,
961
+ "learning_rate": 3.0209111277072444e-05,
962
+ "loss": 1.6593,
963
+ "step": 1590
964
+ },
965
+ {
966
+ "epoch": 1.19,
967
+ "learning_rate": 3.0084640278815036e-05,
968
+ "loss": 1.6248,
969
+ "step": 1600
970
+ },
971
+ {
972
+ "epoch": 1.2,
973
+ "learning_rate": 2.996016928055763e-05,
974
+ "loss": 1.6827,
975
+ "step": 1610
976
+ },
977
+ {
978
+ "epoch": 1.21,
979
+ "learning_rate": 2.9835698282300222e-05,
980
+ "loss": 1.7054,
981
+ "step": 1620
982
+ },
983
+ {
984
+ "epoch": 1.22,
985
+ "learning_rate": 2.971122728404282e-05,
986
+ "loss": 1.6518,
987
+ "step": 1630
988
+ },
989
+ {
990
+ "epoch": 1.22,
991
+ "learning_rate": 2.9586756285785412e-05,
992
+ "loss": 1.6795,
993
+ "step": 1640
994
+ },
995
+ {
996
+ "epoch": 1.23,
997
+ "learning_rate": 2.9462285287528007e-05,
998
+ "loss": 1.6286,
999
+ "step": 1650
1000
+ },
1001
+ {
1002
+ "epoch": 1.24,
1003
+ "learning_rate": 2.93378142892706e-05,
1004
+ "loss": 1.6599,
1005
+ "step": 1660
1006
+ },
1007
+ {
1008
+ "epoch": 1.25,
1009
+ "learning_rate": 2.9213343291013197e-05,
1010
+ "loss": 1.6902,
1011
+ "step": 1670
1012
+ },
1013
+ {
1014
+ "epoch": 1.25,
1015
+ "learning_rate": 2.9088872292755788e-05,
1016
+ "loss": 1.6551,
1017
+ "step": 1680
1018
+ },
1019
+ {
1020
+ "epoch": 1.26,
1021
+ "learning_rate": 2.8964401294498383e-05,
1022
+ "loss": 1.6493,
1023
+ "step": 1690
1024
+ },
1025
+ {
1026
+ "epoch": 1.27,
1027
+ "learning_rate": 2.8839930296240975e-05,
1028
+ "loss": 1.7094,
1029
+ "step": 1700
1030
+ },
1031
+ {
1032
+ "epoch": 1.28,
1033
+ "learning_rate": 2.8715459297983573e-05,
1034
+ "loss": 1.6714,
1035
+ "step": 1710
1036
+ },
1037
+ {
1038
+ "epoch": 1.28,
1039
+ "learning_rate": 2.859098829972616e-05,
1040
+ "loss": 1.6296,
1041
+ "step": 1720
1042
+ },
1043
+ {
1044
+ "epoch": 1.29,
1045
+ "learning_rate": 2.846651730146876e-05,
1046
+ "loss": 1.6882,
1047
+ "step": 1730
1048
+ },
1049
+ {
1050
+ "epoch": 1.3,
1051
+ "learning_rate": 2.834204630321135e-05,
1052
+ "loss": 1.6493,
1053
+ "step": 1740
1054
+ },
1055
+ {
1056
+ "epoch": 1.31,
1057
+ "learning_rate": 2.8217575304953945e-05,
1058
+ "loss": 1.6489,
1059
+ "step": 1750
1060
+ },
1061
+ {
1062
+ "epoch": 1.31,
1063
+ "learning_rate": 2.8093104306696544e-05,
1064
+ "loss": 1.6683,
1065
+ "step": 1760
1066
+ },
1067
+ {
1068
+ "epoch": 1.32,
1069
+ "learning_rate": 2.7968633308439135e-05,
1070
+ "loss": 1.6988,
1071
+ "step": 1770
1072
+ },
1073
+ {
1074
+ "epoch": 1.33,
1075
+ "learning_rate": 2.784416231018173e-05,
1076
+ "loss": 1.648,
1077
+ "step": 1780
1078
+ },
1079
+ {
1080
+ "epoch": 1.34,
1081
+ "learning_rate": 2.771969131192432e-05,
1082
+ "loss": 1.6655,
1083
+ "step": 1790
1084
+ },
1085
+ {
1086
+ "epoch": 1.34,
1087
+ "learning_rate": 2.759522031366692e-05,
1088
+ "loss": 1.6709,
1089
+ "step": 1800
1090
+ },
1091
+ {
1092
+ "epoch": 1.35,
1093
+ "learning_rate": 2.747074931540951e-05,
1094
+ "loss": 1.665,
1095
+ "step": 1810
1096
+ },
1097
+ {
1098
+ "epoch": 1.36,
1099
+ "learning_rate": 2.7346278317152106e-05,
1100
+ "loss": 1.6411,
1101
+ "step": 1820
1102
+ },
1103
+ {
1104
+ "epoch": 1.37,
1105
+ "learning_rate": 2.7221807318894698e-05,
1106
+ "loss": 1.6913,
1107
+ "step": 1830
1108
+ },
1109
+ {
1110
+ "epoch": 1.37,
1111
+ "learning_rate": 2.7097336320637296e-05,
1112
+ "loss": 1.6906,
1113
+ "step": 1840
1114
+ },
1115
+ {
1116
+ "epoch": 1.38,
1117
+ "learning_rate": 2.6972865322379887e-05,
1118
+ "loss": 1.6345,
1119
+ "step": 1850
1120
+ },
1121
+ {
1122
+ "epoch": 1.39,
1123
+ "learning_rate": 2.6848394324122482e-05,
1124
+ "loss": 1.6903,
1125
+ "step": 1860
1126
+ },
1127
+ {
1128
+ "epoch": 1.4,
1129
+ "learning_rate": 2.6723923325865074e-05,
1130
+ "loss": 1.6819,
1131
+ "step": 1870
1132
+ },
1133
+ {
1134
+ "epoch": 1.4,
1135
+ "learning_rate": 2.6599452327607672e-05,
1136
+ "loss": 1.6393,
1137
+ "step": 1880
1138
+ },
1139
+ {
1140
+ "epoch": 1.41,
1141
+ "learning_rate": 2.6474981329350264e-05,
1142
+ "loss": 1.6725,
1143
+ "step": 1890
1144
+ },
1145
+ {
1146
+ "epoch": 1.42,
1147
+ "learning_rate": 2.635051033109286e-05,
1148
+ "loss": 1.6736,
1149
+ "step": 1900
1150
+ },
1151
+ {
1152
+ "epoch": 1.43,
1153
+ "learning_rate": 2.622603933283545e-05,
1154
+ "loss": 1.6734,
1155
+ "step": 1910
1156
+ },
1157
+ {
1158
+ "epoch": 1.43,
1159
+ "learning_rate": 2.6101568334578048e-05,
1160
+ "loss": 1.6761,
1161
+ "step": 1920
1162
+ },
1163
+ {
1164
+ "epoch": 1.44,
1165
+ "learning_rate": 2.597709733632064e-05,
1166
+ "loss": 1.6842,
1167
+ "step": 1930
1168
+ },
1169
+ {
1170
+ "epoch": 1.45,
1171
+ "learning_rate": 2.5852626338063234e-05,
1172
+ "loss": 1.7064,
1173
+ "step": 1940
1174
+ },
1175
+ {
1176
+ "epoch": 1.46,
1177
+ "learning_rate": 2.5728155339805826e-05,
1178
+ "loss": 1.7036,
1179
+ "step": 1950
1180
+ },
1181
+ {
1182
+ "epoch": 1.46,
1183
+ "learning_rate": 2.5603684341548424e-05,
1184
+ "loss": 1.6563,
1185
+ "step": 1960
1186
+ },
1187
+ {
1188
+ "epoch": 1.47,
1189
+ "learning_rate": 2.5479213343291016e-05,
1190
+ "loss": 1.6796,
1191
+ "step": 1970
1192
+ },
1193
+ {
1194
+ "epoch": 1.48,
1195
+ "learning_rate": 2.535474234503361e-05,
1196
+ "loss": 1.6469,
1197
+ "step": 1980
1198
+ },
1199
+ {
1200
+ "epoch": 1.49,
1201
+ "learning_rate": 2.5230271346776202e-05,
1202
+ "loss": 1.6681,
1203
+ "step": 1990
1204
+ },
1205
+ {
1206
+ "epoch": 1.49,
1207
+ "learning_rate": 2.51058003485188e-05,
1208
+ "loss": 1.6632,
1209
+ "step": 2000
1210
+ },
1211
+ {
1212
+ "epoch": 1.5,
1213
+ "learning_rate": 2.498132935026139e-05,
1214
+ "loss": 1.6799,
1215
+ "step": 2010
1216
+ },
1217
+ {
1218
+ "epoch": 1.51,
1219
+ "learning_rate": 2.4856858352003983e-05,
1220
+ "loss": 1.6811,
1221
+ "step": 2020
1222
+ },
1223
+ {
1224
+ "epoch": 1.52,
1225
+ "learning_rate": 2.4732387353746578e-05,
1226
+ "loss": 1.6365,
1227
+ "step": 2030
1228
+ },
1229
+ {
1230
+ "epoch": 1.52,
1231
+ "learning_rate": 2.4607916355489173e-05,
1232
+ "loss": 1.6773,
1233
+ "step": 2040
1234
+ },
1235
+ {
1236
+ "epoch": 1.53,
1237
+ "learning_rate": 2.4483445357231764e-05,
1238
+ "loss": 1.6194,
1239
+ "step": 2050
1240
+ },
1241
+ {
1242
+ "epoch": 1.54,
1243
+ "learning_rate": 2.435897435897436e-05,
1244
+ "loss": 1.6865,
1245
+ "step": 2060
1246
+ },
1247
+ {
1248
+ "epoch": 1.55,
1249
+ "learning_rate": 2.4234503360716954e-05,
1250
+ "loss": 1.6552,
1251
+ "step": 2070
1252
+ },
1253
+ {
1254
+ "epoch": 1.55,
1255
+ "learning_rate": 2.411003236245955e-05,
1256
+ "loss": 1.6823,
1257
+ "step": 2080
1258
+ },
1259
+ {
1260
+ "epoch": 1.56,
1261
+ "learning_rate": 2.398556136420214e-05,
1262
+ "loss": 1.6611,
1263
+ "step": 2090
1264
+ },
1265
+ {
1266
+ "epoch": 1.57,
1267
+ "learning_rate": 2.3861090365944735e-05,
1268
+ "loss": 1.6662,
1269
+ "step": 2100
1270
+ },
1271
+ {
1272
+ "epoch": 1.58,
1273
+ "learning_rate": 2.373661936768733e-05,
1274
+ "loss": 1.6898,
1275
+ "step": 2110
1276
+ },
1277
+ {
1278
+ "epoch": 1.58,
1279
+ "learning_rate": 2.3612148369429922e-05,
1280
+ "loss": 1.6474,
1281
+ "step": 2120
1282
+ },
1283
+ {
1284
+ "epoch": 1.59,
1285
+ "learning_rate": 2.3487677371172517e-05,
1286
+ "loss": 1.6635,
1287
+ "step": 2130
1288
+ },
1289
+ {
1290
+ "epoch": 1.6,
1291
+ "learning_rate": 2.336320637291511e-05,
1292
+ "loss": 1.6274,
1293
+ "step": 2140
1294
+ },
1295
+ {
1296
+ "epoch": 1.61,
1297
+ "learning_rate": 2.3238735374657706e-05,
1298
+ "loss": 1.6794,
1299
+ "step": 2150
1300
+ },
1301
+ {
1302
+ "epoch": 1.61,
1303
+ "learning_rate": 2.3114264376400298e-05,
1304
+ "loss": 1.6411,
1305
+ "step": 2160
1306
+ },
1307
+ {
1308
+ "epoch": 1.62,
1309
+ "learning_rate": 2.2989793378142893e-05,
1310
+ "loss": 1.6793,
1311
+ "step": 2170
1312
+ },
1313
+ {
1314
+ "epoch": 1.63,
1315
+ "learning_rate": 2.2865322379885488e-05,
1316
+ "loss": 1.6772,
1317
+ "step": 2180
1318
+ },
1319
+ {
1320
+ "epoch": 1.64,
1321
+ "learning_rate": 2.2740851381628083e-05,
1322
+ "loss": 1.6425,
1323
+ "step": 2190
1324
+ },
1325
+ {
1326
+ "epoch": 1.64,
1327
+ "learning_rate": 2.2616380383370674e-05,
1328
+ "loss": 1.6614,
1329
+ "step": 2200
1330
+ },
1331
+ {
1332
+ "epoch": 1.65,
1333
+ "learning_rate": 2.249190938511327e-05,
1334
+ "loss": 1.65,
1335
+ "step": 2210
1336
+ },
1337
+ {
1338
+ "epoch": 1.66,
1339
+ "learning_rate": 2.2367438386855864e-05,
1340
+ "loss": 1.685,
1341
+ "step": 2220
1342
+ },
1343
+ {
1344
+ "epoch": 1.67,
1345
+ "learning_rate": 2.2242967388598455e-05,
1346
+ "loss": 1.6808,
1347
+ "step": 2230
1348
+ },
1349
+ {
1350
+ "epoch": 1.67,
1351
+ "learning_rate": 2.211849639034105e-05,
1352
+ "loss": 1.6612,
1353
+ "step": 2240
1354
+ },
1355
+ {
1356
+ "epoch": 1.68,
1357
+ "learning_rate": 2.1994025392083645e-05,
1358
+ "loss": 1.6631,
1359
+ "step": 2250
1360
+ },
1361
+ {
1362
+ "epoch": 1.69,
1363
+ "learning_rate": 2.186955439382624e-05,
1364
+ "loss": 1.6675,
1365
+ "step": 2260
1366
+ },
1367
+ {
1368
+ "epoch": 1.7,
1369
+ "learning_rate": 2.174508339556883e-05,
1370
+ "loss": 1.6532,
1371
+ "step": 2270
1372
+ },
1373
+ {
1374
+ "epoch": 1.7,
1375
+ "learning_rate": 2.1620612397311426e-05,
1376
+ "loss": 1.7217,
1377
+ "step": 2280
1378
+ },
1379
+ {
1380
+ "epoch": 1.71,
1381
+ "learning_rate": 2.149614139905402e-05,
1382
+ "loss": 1.6697,
1383
+ "step": 2290
1384
+ },
1385
+ {
1386
+ "epoch": 1.72,
1387
+ "learning_rate": 2.1371670400796616e-05,
1388
+ "loss": 1.634,
1389
+ "step": 2300
1390
+ },
1391
+ {
1392
+ "epoch": 1.73,
1393
+ "learning_rate": 2.1247199402539207e-05,
1394
+ "loss": 1.6968,
1395
+ "step": 2310
1396
+ },
1397
+ {
1398
+ "epoch": 1.73,
1399
+ "learning_rate": 2.1122728404281802e-05,
1400
+ "loss": 1.6925,
1401
+ "step": 2320
1402
+ },
1403
+ {
1404
+ "epoch": 1.74,
1405
+ "learning_rate": 2.0998257406024397e-05,
1406
+ "loss": 1.6455,
1407
+ "step": 2330
1408
+ },
1409
+ {
1410
+ "epoch": 1.75,
1411
+ "learning_rate": 2.0873786407766992e-05,
1412
+ "loss": 1.6365,
1413
+ "step": 2340
1414
+ },
1415
+ {
1416
+ "epoch": 1.76,
1417
+ "learning_rate": 2.0749315409509583e-05,
1418
+ "loss": 1.6813,
1419
+ "step": 2350
1420
+ },
1421
+ {
1422
+ "epoch": 1.76,
1423
+ "learning_rate": 2.062484441125218e-05,
1424
+ "loss": 1.6725,
1425
+ "step": 2360
1426
+ },
1427
+ {
1428
+ "epoch": 1.77,
1429
+ "learning_rate": 2.0500373412994773e-05,
1430
+ "loss": 1.6718,
1431
+ "step": 2370
1432
+ },
1433
+ {
1434
+ "epoch": 1.78,
1435
+ "learning_rate": 2.0375902414737365e-05,
1436
+ "loss": 1.6954,
1437
+ "step": 2380
1438
+ },
1439
+ {
1440
+ "epoch": 1.78,
1441
+ "learning_rate": 2.025143141647996e-05,
1442
+ "loss": 1.6719,
1443
+ "step": 2390
1444
+ },
1445
+ {
1446
+ "epoch": 1.79,
1447
+ "learning_rate": 2.0126960418222554e-05,
1448
+ "loss": 1.6549,
1449
+ "step": 2400
1450
+ },
1451
+ {
1452
+ "epoch": 1.8,
1453
+ "learning_rate": 2.000248941996515e-05,
1454
+ "loss": 1.6761,
1455
+ "step": 2410
1456
+ },
1457
+ {
1458
+ "epoch": 1.81,
1459
+ "learning_rate": 1.987801842170774e-05,
1460
+ "loss": 1.6582,
1461
+ "step": 2420
1462
+ },
1463
+ {
1464
+ "epoch": 1.81,
1465
+ "learning_rate": 1.9753547423450336e-05,
1466
+ "loss": 1.6202,
1467
+ "step": 2430
1468
+ },
1469
+ {
1470
+ "epoch": 1.82,
1471
+ "learning_rate": 1.962907642519293e-05,
1472
+ "loss": 1.6645,
1473
+ "step": 2440
1474
+ },
1475
+ {
1476
+ "epoch": 1.83,
1477
+ "learning_rate": 1.9504605426935525e-05,
1478
+ "loss": 1.6565,
1479
+ "step": 2450
1480
+ },
1481
+ {
1482
+ "epoch": 1.84,
1483
+ "learning_rate": 1.9380134428678117e-05,
1484
+ "loss": 1.6395,
1485
+ "step": 2460
1486
+ },
1487
+ {
1488
+ "epoch": 1.84,
1489
+ "learning_rate": 1.9255663430420712e-05,
1490
+ "loss": 1.6469,
1491
+ "step": 2470
1492
+ },
1493
+ {
1494
+ "epoch": 1.85,
1495
+ "learning_rate": 1.9131192432163307e-05,
1496
+ "loss": 1.6523,
1497
+ "step": 2480
1498
+ },
1499
+ {
1500
+ "epoch": 1.86,
1501
+ "learning_rate": 1.9006721433905898e-05,
1502
+ "loss": 1.6799,
1503
+ "step": 2490
1504
+ },
1505
+ {
1506
+ "epoch": 1.87,
1507
+ "learning_rate": 1.8882250435648493e-05,
1508
+ "loss": 1.6584,
1509
+ "step": 2500
1510
+ },
1511
+ {
1512
+ "epoch": 1.87,
1513
+ "learning_rate": 1.8757779437391088e-05,
1514
+ "loss": 1.6729,
1515
+ "step": 2510
1516
+ },
1517
+ {
1518
+ "epoch": 1.88,
1519
+ "learning_rate": 1.8633308439133683e-05,
1520
+ "loss": 1.6157,
1521
+ "step": 2520
1522
+ },
1523
+ {
1524
+ "epoch": 1.89,
1525
+ "learning_rate": 1.8508837440876274e-05,
1526
+ "loss": 1.7094,
1527
+ "step": 2530
1528
+ },
1529
+ {
1530
+ "epoch": 1.9,
1531
+ "learning_rate": 1.838436644261887e-05,
1532
+ "loss": 1.68,
1533
+ "step": 2540
1534
+ },
1535
+ {
1536
+ "epoch": 1.9,
1537
+ "learning_rate": 1.8259895444361464e-05,
1538
+ "loss": 1.7009,
1539
+ "step": 2550
1540
+ },
1541
+ {
1542
+ "epoch": 1.91,
1543
+ "learning_rate": 1.813542444610406e-05,
1544
+ "loss": 1.6931,
1545
+ "step": 2560
1546
+ },
1547
+ {
1548
+ "epoch": 1.92,
1549
+ "learning_rate": 1.801095344784665e-05,
1550
+ "loss": 1.7001,
1551
+ "step": 2570
1552
+ },
1553
+ {
1554
+ "epoch": 1.93,
1555
+ "learning_rate": 1.7886482449589245e-05,
1556
+ "loss": 1.6847,
1557
+ "step": 2580
1558
+ },
1559
+ {
1560
+ "epoch": 1.93,
1561
+ "learning_rate": 1.776201145133184e-05,
1562
+ "loss": 1.653,
1563
+ "step": 2590
1564
+ },
1565
+ {
1566
+ "epoch": 1.94,
1567
+ "learning_rate": 1.763754045307443e-05,
1568
+ "loss": 1.6386,
1569
+ "step": 2600
1570
+ },
1571
+ {
1572
+ "epoch": 1.95,
1573
+ "learning_rate": 1.7513069454817026e-05,
1574
+ "loss": 1.6832,
1575
+ "step": 2610
1576
+ },
1577
+ {
1578
+ "epoch": 1.96,
1579
+ "learning_rate": 1.7388598456559625e-05,
1580
+ "loss": 1.6853,
1581
+ "step": 2620
1582
+ },
1583
+ {
1584
+ "epoch": 1.96,
1585
+ "learning_rate": 1.7264127458302216e-05,
1586
+ "loss": 1.6876,
1587
+ "step": 2630
1588
+ },
1589
+ {
1590
+ "epoch": 1.97,
1591
+ "learning_rate": 1.713965646004481e-05,
1592
+ "loss": 1.6878,
1593
+ "step": 2640
1594
+ },
1595
+ {
1596
+ "epoch": 1.98,
1597
+ "learning_rate": 1.7015185461787406e-05,
1598
+ "loss": 1.7067,
1599
+ "step": 2650
1600
+ },
1601
+ {
1602
+ "epoch": 1.99,
1603
+ "learning_rate": 1.689071446353e-05,
1604
+ "loss": 1.6736,
1605
+ "step": 2660
1606
+ },
1607
+ {
1608
+ "epoch": 1.99,
1609
+ "learning_rate": 1.6766243465272592e-05,
1610
+ "loss": 1.6536,
1611
+ "step": 2670
1612
+ },
1613
+ {
1614
+ "epoch": 2.0,
1615
+ "learning_rate": 1.6641772467015187e-05,
1616
+ "loss": 1.6806,
1617
+ "step": 2680
1618
+ },
1619
+ {
1620
+ "epoch": 2.01,
1621
+ "learning_rate": 1.6517301468757782e-05,
1622
+ "loss": 1.6708,
1623
+ "step": 2690
1624
+ },
1625
+ {
1626
+ "epoch": 2.02,
1627
+ "learning_rate": 1.6392830470500377e-05,
1628
+ "loss": 1.6718,
1629
+ "step": 2700
1630
+ },
1631
+ {
1632
+ "epoch": 2.02,
1633
+ "learning_rate": 1.626835947224297e-05,
1634
+ "loss": 1.6829,
1635
+ "step": 2710
1636
+ },
1637
+ {
1638
+ "epoch": 2.03,
1639
+ "learning_rate": 1.6143888473985563e-05,
1640
+ "loss": 1.7071,
1641
+ "step": 2720
1642
+ },
1643
+ {
1644
+ "epoch": 2.04,
1645
+ "learning_rate": 1.6019417475728158e-05,
1646
+ "loss": 1.6583,
1647
+ "step": 2730
1648
+ },
1649
+ {
1650
+ "epoch": 2.05,
1651
+ "learning_rate": 1.5894946477470753e-05,
1652
+ "loss": 1.6634,
1653
+ "step": 2740
1654
+ },
1655
+ {
1656
+ "epoch": 2.05,
1657
+ "learning_rate": 1.5770475479213344e-05,
1658
+ "loss": 1.6614,
1659
+ "step": 2750
1660
+ },
1661
+ {
1662
+ "epoch": 2.06,
1663
+ "learning_rate": 1.564600448095594e-05,
1664
+ "loss": 1.6689,
1665
+ "step": 2760
1666
+ },
1667
+ {
1668
+ "epoch": 2.07,
1669
+ "learning_rate": 1.5521533482698534e-05,
1670
+ "loss": 1.6578,
1671
+ "step": 2770
1672
+ },
1673
+ {
1674
+ "epoch": 2.08,
1675
+ "learning_rate": 1.5397062484441126e-05,
1676
+ "loss": 1.6791,
1677
+ "step": 2780
1678
+ },
1679
+ {
1680
+ "epoch": 2.08,
1681
+ "learning_rate": 1.527259148618372e-05,
1682
+ "loss": 1.6402,
1683
+ "step": 2790
1684
+ },
1685
+ {
1686
+ "epoch": 2.09,
1687
+ "learning_rate": 1.5148120487926315e-05,
1688
+ "loss": 1.6574,
1689
+ "step": 2800
1690
+ },
1691
+ {
1692
+ "epoch": 2.1,
1693
+ "learning_rate": 1.5023649489668909e-05,
1694
+ "loss": 1.6598,
1695
+ "step": 2810
1696
+ },
1697
+ {
1698
+ "epoch": 2.11,
1699
+ "learning_rate": 1.4899178491411503e-05,
1700
+ "loss": 1.6827,
1701
+ "step": 2820
1702
+ },
1703
+ {
1704
+ "epoch": 2.11,
1705
+ "learning_rate": 1.4774707493154097e-05,
1706
+ "loss": 1.6471,
1707
+ "step": 2830
1708
+ },
1709
+ {
1710
+ "epoch": 2.12,
1711
+ "learning_rate": 1.4650236494896691e-05,
1712
+ "loss": 1.6398,
1713
+ "step": 2840
1714
+ },
1715
+ {
1716
+ "epoch": 2.13,
1717
+ "learning_rate": 1.4525765496639285e-05,
1718
+ "loss": 1.6557,
1719
+ "step": 2850
1720
+ },
1721
+ {
1722
+ "epoch": 2.14,
1723
+ "learning_rate": 1.440129449838188e-05,
1724
+ "loss": 1.6614,
1725
+ "step": 2860
1726
+ },
1727
+ {
1728
+ "epoch": 2.14,
1729
+ "learning_rate": 1.4276823500124473e-05,
1730
+ "loss": 1.6619,
1731
+ "step": 2870
1732
+ },
1733
+ {
1734
+ "epoch": 2.15,
1735
+ "learning_rate": 1.4152352501867066e-05,
1736
+ "loss": 1.6887,
1737
+ "step": 2880
1738
+ },
1739
+ {
1740
+ "epoch": 2.16,
1741
+ "learning_rate": 1.402788150360966e-05,
1742
+ "loss": 1.6815,
1743
+ "step": 2890
1744
+ },
1745
+ {
1746
+ "epoch": 2.17,
1747
+ "learning_rate": 1.3903410505352254e-05,
1748
+ "loss": 1.6267,
1749
+ "step": 2900
1750
+ },
1751
+ {
1752
+ "epoch": 2.17,
1753
+ "learning_rate": 1.3778939507094849e-05,
1754
+ "loss": 1.6938,
1755
+ "step": 2910
1756
+ },
1757
+ {
1758
+ "epoch": 2.18,
1759
+ "learning_rate": 1.3654468508837442e-05,
1760
+ "loss": 1.6796,
1761
+ "step": 2920
1762
+ },
1763
+ {
1764
+ "epoch": 2.19,
1765
+ "learning_rate": 1.3529997510580037e-05,
1766
+ "loss": 1.6544,
1767
+ "step": 2930
1768
+ },
1769
+ {
1770
+ "epoch": 2.2,
1771
+ "learning_rate": 1.340552651232263e-05,
1772
+ "loss": 1.6497,
1773
+ "step": 2940
1774
+ },
1775
+ {
1776
+ "epoch": 2.2,
1777
+ "learning_rate": 1.3281055514065225e-05,
1778
+ "loss": 1.6397,
1779
+ "step": 2950
1780
+ },
1781
+ {
1782
+ "epoch": 2.21,
1783
+ "learning_rate": 1.3156584515807818e-05,
1784
+ "loss": 1.6788,
1785
+ "step": 2960
1786
+ },
1787
+ {
1788
+ "epoch": 2.22,
1789
+ "learning_rate": 1.3032113517550413e-05,
1790
+ "loss": 1.6675,
1791
+ "step": 2970
1792
+ },
1793
+ {
1794
+ "epoch": 2.23,
1795
+ "learning_rate": 1.2907642519293006e-05,
1796
+ "loss": 1.6429,
1797
+ "step": 2980
1798
+ },
1799
+ {
1800
+ "epoch": 2.23,
1801
+ "learning_rate": 1.27831715210356e-05,
1802
+ "loss": 1.6866,
1803
+ "step": 2990
1804
+ },
1805
+ {
1806
+ "epoch": 2.24,
1807
+ "learning_rate": 1.2658700522778194e-05,
1808
+ "loss": 1.6773,
1809
+ "step": 3000
1810
+ }
1811
+ ],
1812
+ "logging_steps": 10,
1813
+ "max_steps": 4017,
1814
+ "num_input_tokens_seen": 0,
1815
+ "num_train_epochs": 3,
1816
+ "save_steps": 500,
1817
+ "total_flos": 1.89607446577152e+18,
1818
+ "train_batch_size": 4,
1819
+ "trial_name": null,
1820
+ "trial_params": null
1821
+ }
119_Llama-2-13b-hf/checkpoint-3000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:95fdd19e2ed6f33177f737bfa3ccdb7380f84c750beb151600638af981e7df02
3
+ size 4283
119_Llama-2-13b-hf/checkpoint-3500/README.md ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+ ### Framework versions
7
+
8
+
9
+ - PEFT 0.4.0
119_Llama-2-13b-hf/checkpoint-3500/adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "meta-llama/Llama-2-13b-hf",
4
+ "bias": "none",
5
+ "fan_in_fan_out": false,
6
+ "inference_mode": true,
7
+ "init_lora_weights": true,
8
+ "layers_pattern": null,
9
+ "layers_to_transform": null,
10
+ "lora_alpha": 32,
11
+ "lora_dropout": 0.05,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 8,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "q_proj",
18
+ "v_proj"
19
+ ],
20
+ "task_type": "CAUSAL_LM"
21
+ }
119_Llama-2-13b-hf/checkpoint-3500/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:701a80cb3cd427ea37bf892408da47a0e8d4229104b627c78cc575142436cd26
3
+ size 26235704
119_Llama-2-13b-hf/checkpoint-3500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:72ff829c7ee1f05dac05e1a07a2224905f0ababb1a83570c02a48e44889281a0
3
+ size 52567237
119_Llama-2-13b-hf/checkpoint-3500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fd2119dc61084423bec439b9f1e4542b38aac5daaaf06856866e02395fd574ab
3
+ size 14575
119_Llama-2-13b-hf/checkpoint-3500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ab057d7f09fea26670ea9c30bf9c4685f5ff1b7c73eb68cd54f8088f56cad01b
3
+ size 627
119_Llama-2-13b-hf/checkpoint-3500/trainer_state.json ADDED
@@ -0,0 +1,2121 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.6138909634055265,
5
+ "eval_steps": 500,
6
+ "global_step": 3500,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.01,
13
+ "learning_rate": 4.98755290017426e-05,
14
+ "loss": 1.7891,
15
+ "step": 10
16
+ },
17
+ {
18
+ "epoch": 0.01,
19
+ "learning_rate": 4.975105800348519e-05,
20
+ "loss": 1.7869,
21
+ "step": 20
22
+ },
23
+ {
24
+ "epoch": 0.02,
25
+ "learning_rate": 4.962658700522779e-05,
26
+ "loss": 1.7868,
27
+ "step": 30
28
+ },
29
+ {
30
+ "epoch": 0.03,
31
+ "learning_rate": 4.9502116006970375e-05,
32
+ "loss": 1.7547,
33
+ "step": 40
34
+ },
35
+ {
36
+ "epoch": 0.04,
37
+ "learning_rate": 4.937764500871298e-05,
38
+ "loss": 1.6772,
39
+ "step": 50
40
+ },
41
+ {
42
+ "epoch": 0.04,
43
+ "learning_rate": 4.9253174010455565e-05,
44
+ "loss": 1.7076,
45
+ "step": 60
46
+ },
47
+ {
48
+ "epoch": 0.05,
49
+ "learning_rate": 4.912870301219816e-05,
50
+ "loss": 1.6948,
51
+ "step": 70
52
+ },
53
+ {
54
+ "epoch": 0.06,
55
+ "learning_rate": 4.9004232013940755e-05,
56
+ "loss": 1.7135,
57
+ "step": 80
58
+ },
59
+ {
60
+ "epoch": 0.07,
61
+ "learning_rate": 4.887976101568335e-05,
62
+ "loss": 1.7007,
63
+ "step": 90
64
+ },
65
+ {
66
+ "epoch": 0.07,
67
+ "learning_rate": 4.8755290017425944e-05,
68
+ "loss": 1.7247,
69
+ "step": 100
70
+ },
71
+ {
72
+ "epoch": 0.08,
73
+ "learning_rate": 4.863081901916854e-05,
74
+ "loss": 1.719,
75
+ "step": 110
76
+ },
77
+ {
78
+ "epoch": 0.09,
79
+ "learning_rate": 4.850634802091113e-05,
80
+ "loss": 1.7209,
81
+ "step": 120
82
+ },
83
+ {
84
+ "epoch": 0.1,
85
+ "learning_rate": 4.838187702265373e-05,
86
+ "loss": 1.7354,
87
+ "step": 130
88
+ },
89
+ {
90
+ "epoch": 0.1,
91
+ "learning_rate": 4.825740602439632e-05,
92
+ "loss": 1.7025,
93
+ "step": 140
94
+ },
95
+ {
96
+ "epoch": 0.11,
97
+ "learning_rate": 4.813293502613891e-05,
98
+ "loss": 1.6664,
99
+ "step": 150
100
+ },
101
+ {
102
+ "epoch": 0.12,
103
+ "learning_rate": 4.800846402788151e-05,
104
+ "loss": 1.6851,
105
+ "step": 160
106
+ },
107
+ {
108
+ "epoch": 0.13,
109
+ "learning_rate": 4.78839930296241e-05,
110
+ "loss": 1.6749,
111
+ "step": 170
112
+ },
113
+ {
114
+ "epoch": 0.13,
115
+ "learning_rate": 4.7759522031366697e-05,
116
+ "loss": 1.6887,
117
+ "step": 180
118
+ },
119
+ {
120
+ "epoch": 0.14,
121
+ "learning_rate": 4.763505103310929e-05,
122
+ "loss": 1.6571,
123
+ "step": 190
124
+ },
125
+ {
126
+ "epoch": 0.15,
127
+ "learning_rate": 4.751058003485188e-05,
128
+ "loss": 1.7136,
129
+ "step": 200
130
+ },
131
+ {
132
+ "epoch": 0.16,
133
+ "learning_rate": 4.738610903659448e-05,
134
+ "loss": 1.7098,
135
+ "step": 210
136
+ },
137
+ {
138
+ "epoch": 0.16,
139
+ "learning_rate": 4.726163803833707e-05,
140
+ "loss": 1.7285,
141
+ "step": 220
142
+ },
143
+ {
144
+ "epoch": 0.17,
145
+ "learning_rate": 4.7137167040079664e-05,
146
+ "loss": 1.7038,
147
+ "step": 230
148
+ },
149
+ {
150
+ "epoch": 0.18,
151
+ "learning_rate": 4.701269604182226e-05,
152
+ "loss": 1.692,
153
+ "step": 240
154
+ },
155
+ {
156
+ "epoch": 0.19,
157
+ "learning_rate": 4.6888225043564854e-05,
158
+ "loss": 1.6825,
159
+ "step": 250
160
+ },
161
+ {
162
+ "epoch": 0.19,
163
+ "learning_rate": 4.676375404530744e-05,
164
+ "loss": 1.7254,
165
+ "step": 260
166
+ },
167
+ {
168
+ "epoch": 0.2,
169
+ "learning_rate": 4.6639283047050044e-05,
170
+ "loss": 1.6765,
171
+ "step": 270
172
+ },
173
+ {
174
+ "epoch": 0.21,
175
+ "learning_rate": 4.651481204879263e-05,
176
+ "loss": 1.6796,
177
+ "step": 280
178
+ },
179
+ {
180
+ "epoch": 0.22,
181
+ "learning_rate": 4.6390341050535227e-05,
182
+ "loss": 1.7302,
183
+ "step": 290
184
+ },
185
+ {
186
+ "epoch": 0.22,
187
+ "learning_rate": 4.626587005227782e-05,
188
+ "loss": 1.7469,
189
+ "step": 300
190
+ },
191
+ {
192
+ "epoch": 0.23,
193
+ "learning_rate": 4.6141399054020416e-05,
194
+ "loss": 1.6345,
195
+ "step": 310
196
+ },
197
+ {
198
+ "epoch": 0.24,
199
+ "learning_rate": 4.601692805576301e-05,
200
+ "loss": 1.6761,
201
+ "step": 320
202
+ },
203
+ {
204
+ "epoch": 0.25,
205
+ "learning_rate": 4.5892457057505606e-05,
206
+ "loss": 1.6602,
207
+ "step": 330
208
+ },
209
+ {
210
+ "epoch": 0.25,
211
+ "learning_rate": 4.5767986059248194e-05,
212
+ "loss": 1.6637,
213
+ "step": 340
214
+ },
215
+ {
216
+ "epoch": 0.26,
217
+ "learning_rate": 4.5643515060990796e-05,
218
+ "loss": 1.6953,
219
+ "step": 350
220
+ },
221
+ {
222
+ "epoch": 0.27,
223
+ "learning_rate": 4.5519044062733384e-05,
224
+ "loss": 1.6829,
225
+ "step": 360
226
+ },
227
+ {
228
+ "epoch": 0.28,
229
+ "learning_rate": 4.539457306447598e-05,
230
+ "loss": 1.657,
231
+ "step": 370
232
+ },
233
+ {
234
+ "epoch": 0.28,
235
+ "learning_rate": 4.5270102066218574e-05,
236
+ "loss": 1.647,
237
+ "step": 380
238
+ },
239
+ {
240
+ "epoch": 0.29,
241
+ "learning_rate": 4.514563106796117e-05,
242
+ "loss": 1.6854,
243
+ "step": 390
244
+ },
245
+ {
246
+ "epoch": 0.3,
247
+ "learning_rate": 4.502116006970376e-05,
248
+ "loss": 1.6737,
249
+ "step": 400
250
+ },
251
+ {
252
+ "epoch": 0.31,
253
+ "learning_rate": 4.489668907144636e-05,
254
+ "loss": 1.6938,
255
+ "step": 410
256
+ },
257
+ {
258
+ "epoch": 0.31,
259
+ "learning_rate": 4.4772218073188946e-05,
260
+ "loss": 1.6721,
261
+ "step": 420
262
+ },
263
+ {
264
+ "epoch": 0.32,
265
+ "learning_rate": 4.464774707493155e-05,
266
+ "loss": 1.6683,
267
+ "step": 430
268
+ },
269
+ {
270
+ "epoch": 0.33,
271
+ "learning_rate": 4.4523276076674136e-05,
272
+ "loss": 1.6229,
273
+ "step": 440
274
+ },
275
+ {
276
+ "epoch": 0.34,
277
+ "learning_rate": 4.439880507841673e-05,
278
+ "loss": 1.6593,
279
+ "step": 450
280
+ },
281
+ {
282
+ "epoch": 0.34,
283
+ "learning_rate": 4.4274334080159326e-05,
284
+ "loss": 1.6844,
285
+ "step": 460
286
+ },
287
+ {
288
+ "epoch": 0.35,
289
+ "learning_rate": 4.414986308190192e-05,
290
+ "loss": 1.6754,
291
+ "step": 470
292
+ },
293
+ {
294
+ "epoch": 0.36,
295
+ "learning_rate": 4.4025392083644516e-05,
296
+ "loss": 1.6747,
297
+ "step": 480
298
+ },
299
+ {
300
+ "epoch": 0.37,
301
+ "learning_rate": 4.390092108538711e-05,
302
+ "loss": 1.6789,
303
+ "step": 490
304
+ },
305
+ {
306
+ "epoch": 0.37,
307
+ "learning_rate": 4.37764500871297e-05,
308
+ "loss": 1.6956,
309
+ "step": 500
310
+ },
311
+ {
312
+ "epoch": 0.38,
313
+ "learning_rate": 4.36519790888723e-05,
314
+ "loss": 1.689,
315
+ "step": 510
316
+ },
317
+ {
318
+ "epoch": 0.39,
319
+ "learning_rate": 4.352750809061489e-05,
320
+ "loss": 1.673,
321
+ "step": 520
322
+ },
323
+ {
324
+ "epoch": 0.4,
325
+ "learning_rate": 4.340303709235748e-05,
326
+ "loss": 1.6968,
327
+ "step": 530
328
+ },
329
+ {
330
+ "epoch": 0.4,
331
+ "learning_rate": 4.327856609410008e-05,
332
+ "loss": 1.6569,
333
+ "step": 540
334
+ },
335
+ {
336
+ "epoch": 0.41,
337
+ "learning_rate": 4.315409509584267e-05,
338
+ "loss": 1.6777,
339
+ "step": 550
340
+ },
341
+ {
342
+ "epoch": 0.42,
343
+ "learning_rate": 4.302962409758526e-05,
344
+ "loss": 1.71,
345
+ "step": 560
346
+ },
347
+ {
348
+ "epoch": 0.43,
349
+ "learning_rate": 4.290515309932786e-05,
350
+ "loss": 1.6943,
351
+ "step": 570
352
+ },
353
+ {
354
+ "epoch": 0.43,
355
+ "learning_rate": 4.278068210107045e-05,
356
+ "loss": 1.6925,
357
+ "step": 580
358
+ },
359
+ {
360
+ "epoch": 0.44,
361
+ "learning_rate": 4.2656211102813046e-05,
362
+ "loss": 1.6628,
363
+ "step": 590
364
+ },
365
+ {
366
+ "epoch": 0.45,
367
+ "learning_rate": 4.253174010455564e-05,
368
+ "loss": 1.7123,
369
+ "step": 600
370
+ },
371
+ {
372
+ "epoch": 0.46,
373
+ "learning_rate": 4.2407269106298235e-05,
374
+ "loss": 1.6772,
375
+ "step": 610
376
+ },
377
+ {
378
+ "epoch": 0.46,
379
+ "learning_rate": 4.228279810804083e-05,
380
+ "loss": 1.6429,
381
+ "step": 620
382
+ },
383
+ {
384
+ "epoch": 0.47,
385
+ "learning_rate": 4.2158327109783425e-05,
386
+ "loss": 1.6472,
387
+ "step": 630
388
+ },
389
+ {
390
+ "epoch": 0.48,
391
+ "learning_rate": 4.203385611152601e-05,
392
+ "loss": 1.6825,
393
+ "step": 640
394
+ },
395
+ {
396
+ "epoch": 0.49,
397
+ "learning_rate": 4.1909385113268615e-05,
398
+ "loss": 1.6825,
399
+ "step": 650
400
+ },
401
+ {
402
+ "epoch": 0.49,
403
+ "learning_rate": 4.17849141150112e-05,
404
+ "loss": 1.6818,
405
+ "step": 660
406
+ },
407
+ {
408
+ "epoch": 0.5,
409
+ "learning_rate": 4.16604431167538e-05,
410
+ "loss": 1.7325,
411
+ "step": 670
412
+ },
413
+ {
414
+ "epoch": 0.51,
415
+ "learning_rate": 4.153597211849639e-05,
416
+ "loss": 1.7008,
417
+ "step": 680
418
+ },
419
+ {
420
+ "epoch": 0.52,
421
+ "learning_rate": 4.141150112023899e-05,
422
+ "loss": 1.655,
423
+ "step": 690
424
+ },
425
+ {
426
+ "epoch": 0.52,
427
+ "learning_rate": 4.128703012198158e-05,
428
+ "loss": 1.6718,
429
+ "step": 700
430
+ },
431
+ {
432
+ "epoch": 0.53,
433
+ "learning_rate": 4.116255912372418e-05,
434
+ "loss": 1.6412,
435
+ "step": 710
436
+ },
437
+ {
438
+ "epoch": 0.54,
439
+ "learning_rate": 4.1038088125466765e-05,
440
+ "loss": 1.7091,
441
+ "step": 720
442
+ },
443
+ {
444
+ "epoch": 0.55,
445
+ "learning_rate": 4.091361712720937e-05,
446
+ "loss": 1.726,
447
+ "step": 730
448
+ },
449
+ {
450
+ "epoch": 0.55,
451
+ "learning_rate": 4.0789146128951955e-05,
452
+ "loss": 1.6655,
453
+ "step": 740
454
+ },
455
+ {
456
+ "epoch": 0.56,
457
+ "learning_rate": 4.066467513069455e-05,
458
+ "loss": 1.7242,
459
+ "step": 750
460
+ },
461
+ {
462
+ "epoch": 0.57,
463
+ "learning_rate": 4.0540204132437145e-05,
464
+ "loss": 1.6714,
465
+ "step": 760
466
+ },
467
+ {
468
+ "epoch": 0.58,
469
+ "learning_rate": 4.041573313417974e-05,
470
+ "loss": 1.6646,
471
+ "step": 770
472
+ },
473
+ {
474
+ "epoch": 0.58,
475
+ "learning_rate": 4.029126213592233e-05,
476
+ "loss": 1.6809,
477
+ "step": 780
478
+ },
479
+ {
480
+ "epoch": 0.59,
481
+ "learning_rate": 4.016679113766493e-05,
482
+ "loss": 1.6868,
483
+ "step": 790
484
+ },
485
+ {
486
+ "epoch": 0.6,
487
+ "learning_rate": 4.004232013940752e-05,
488
+ "loss": 1.694,
489
+ "step": 800
490
+ },
491
+ {
492
+ "epoch": 0.6,
493
+ "learning_rate": 3.991784914115011e-05,
494
+ "loss": 1.6801,
495
+ "step": 810
496
+ },
497
+ {
498
+ "epoch": 0.61,
499
+ "learning_rate": 3.979337814289271e-05,
500
+ "loss": 1.6861,
501
+ "step": 820
502
+ },
503
+ {
504
+ "epoch": 0.62,
505
+ "learning_rate": 3.96689071446353e-05,
506
+ "loss": 1.6646,
507
+ "step": 830
508
+ },
509
+ {
510
+ "epoch": 0.63,
511
+ "learning_rate": 3.95444361463779e-05,
512
+ "loss": 1.6764,
513
+ "step": 840
514
+ },
515
+ {
516
+ "epoch": 0.63,
517
+ "learning_rate": 3.941996514812049e-05,
518
+ "loss": 1.6821,
519
+ "step": 850
520
+ },
521
+ {
522
+ "epoch": 0.64,
523
+ "learning_rate": 3.929549414986308e-05,
524
+ "loss": 1.6681,
525
+ "step": 860
526
+ },
527
+ {
528
+ "epoch": 0.65,
529
+ "learning_rate": 3.917102315160568e-05,
530
+ "loss": 1.6847,
531
+ "step": 870
532
+ },
533
+ {
534
+ "epoch": 0.66,
535
+ "learning_rate": 3.904655215334827e-05,
536
+ "loss": 1.6084,
537
+ "step": 880
538
+ },
539
+ {
540
+ "epoch": 0.66,
541
+ "learning_rate": 3.8922081155090865e-05,
542
+ "loss": 1.6624,
543
+ "step": 890
544
+ },
545
+ {
546
+ "epoch": 0.67,
547
+ "learning_rate": 3.879761015683346e-05,
548
+ "loss": 1.6619,
549
+ "step": 900
550
+ },
551
+ {
552
+ "epoch": 0.68,
553
+ "learning_rate": 3.8673139158576054e-05,
554
+ "loss": 1.7024,
555
+ "step": 910
556
+ },
557
+ {
558
+ "epoch": 0.69,
559
+ "learning_rate": 3.854866816031865e-05,
560
+ "loss": 1.7301,
561
+ "step": 920
562
+ },
563
+ {
564
+ "epoch": 0.69,
565
+ "learning_rate": 3.8424197162061244e-05,
566
+ "loss": 1.6907,
567
+ "step": 930
568
+ },
569
+ {
570
+ "epoch": 0.7,
571
+ "learning_rate": 3.829972616380383e-05,
572
+ "loss": 1.7046,
573
+ "step": 940
574
+ },
575
+ {
576
+ "epoch": 0.71,
577
+ "learning_rate": 3.8175255165546434e-05,
578
+ "loss": 1.6661,
579
+ "step": 950
580
+ },
581
+ {
582
+ "epoch": 0.72,
583
+ "learning_rate": 3.805078416728902e-05,
584
+ "loss": 1.6981,
585
+ "step": 960
586
+ },
587
+ {
588
+ "epoch": 0.72,
589
+ "learning_rate": 3.792631316903162e-05,
590
+ "loss": 1.662,
591
+ "step": 970
592
+ },
593
+ {
594
+ "epoch": 0.73,
595
+ "learning_rate": 3.780184217077421e-05,
596
+ "loss": 1.7073,
597
+ "step": 980
598
+ },
599
+ {
600
+ "epoch": 0.74,
601
+ "learning_rate": 3.7677371172516806e-05,
602
+ "loss": 1.6488,
603
+ "step": 990
604
+ },
605
+ {
606
+ "epoch": 0.75,
607
+ "learning_rate": 3.7552900174259395e-05,
608
+ "loss": 1.6708,
609
+ "step": 1000
610
+ },
611
+ {
612
+ "epoch": 0.75,
613
+ "learning_rate": 3.7428429176001996e-05,
614
+ "loss": 1.7045,
615
+ "step": 1010
616
+ },
617
+ {
618
+ "epoch": 0.76,
619
+ "learning_rate": 3.7303958177744584e-05,
620
+ "loss": 1.6842,
621
+ "step": 1020
622
+ },
623
+ {
624
+ "epoch": 0.77,
625
+ "learning_rate": 3.717948717948718e-05,
626
+ "loss": 1.6682,
627
+ "step": 1030
628
+ },
629
+ {
630
+ "epoch": 0.78,
631
+ "learning_rate": 3.7055016181229774e-05,
632
+ "loss": 1.636,
633
+ "step": 1040
634
+ },
635
+ {
636
+ "epoch": 0.78,
637
+ "learning_rate": 3.693054518297237e-05,
638
+ "loss": 1.6801,
639
+ "step": 1050
640
+ },
641
+ {
642
+ "epoch": 0.79,
643
+ "learning_rate": 3.6806074184714964e-05,
644
+ "loss": 1.6371,
645
+ "step": 1060
646
+ },
647
+ {
648
+ "epoch": 0.8,
649
+ "learning_rate": 3.668160318645756e-05,
650
+ "loss": 1.6755,
651
+ "step": 1070
652
+ },
653
+ {
654
+ "epoch": 0.81,
655
+ "learning_rate": 3.655713218820015e-05,
656
+ "loss": 1.7046,
657
+ "step": 1080
658
+ },
659
+ {
660
+ "epoch": 0.81,
661
+ "learning_rate": 3.643266118994275e-05,
662
+ "loss": 1.707,
663
+ "step": 1090
664
+ },
665
+ {
666
+ "epoch": 0.82,
667
+ "learning_rate": 3.6308190191685337e-05,
668
+ "loss": 1.6568,
669
+ "step": 1100
670
+ },
671
+ {
672
+ "epoch": 0.83,
673
+ "learning_rate": 3.618371919342793e-05,
674
+ "loss": 1.6447,
675
+ "step": 1110
676
+ },
677
+ {
678
+ "epoch": 0.84,
679
+ "learning_rate": 3.6059248195170526e-05,
680
+ "loss": 1.691,
681
+ "step": 1120
682
+ },
683
+ {
684
+ "epoch": 0.84,
685
+ "learning_rate": 3.593477719691312e-05,
686
+ "loss": 1.6747,
687
+ "step": 1130
688
+ },
689
+ {
690
+ "epoch": 0.85,
691
+ "learning_rate": 3.5810306198655716e-05,
692
+ "loss": 1.656,
693
+ "step": 1140
694
+ },
695
+ {
696
+ "epoch": 0.86,
697
+ "learning_rate": 3.568583520039831e-05,
698
+ "loss": 1.6485,
699
+ "step": 1150
700
+ },
701
+ {
702
+ "epoch": 0.87,
703
+ "learning_rate": 3.55613642021409e-05,
704
+ "loss": 1.642,
705
+ "step": 1160
706
+ },
707
+ {
708
+ "epoch": 0.87,
709
+ "learning_rate": 3.54368932038835e-05,
710
+ "loss": 1.6891,
711
+ "step": 1170
712
+ },
713
+ {
714
+ "epoch": 0.88,
715
+ "learning_rate": 3.531242220562609e-05,
716
+ "loss": 1.716,
717
+ "step": 1180
718
+ },
719
+ {
720
+ "epoch": 0.89,
721
+ "learning_rate": 3.5187951207368684e-05,
722
+ "loss": 1.6785,
723
+ "step": 1190
724
+ },
725
+ {
726
+ "epoch": 0.9,
727
+ "learning_rate": 3.506348020911128e-05,
728
+ "loss": 1.7053,
729
+ "step": 1200
730
+ },
731
+ {
732
+ "epoch": 0.9,
733
+ "learning_rate": 3.493900921085387e-05,
734
+ "loss": 1.6833,
735
+ "step": 1210
736
+ },
737
+ {
738
+ "epoch": 0.91,
739
+ "learning_rate": 3.481453821259646e-05,
740
+ "loss": 1.7336,
741
+ "step": 1220
742
+ },
743
+ {
744
+ "epoch": 0.92,
745
+ "learning_rate": 3.469006721433906e-05,
746
+ "loss": 1.6496,
747
+ "step": 1230
748
+ },
749
+ {
750
+ "epoch": 0.93,
751
+ "learning_rate": 3.456559621608165e-05,
752
+ "loss": 1.7014,
753
+ "step": 1240
754
+ },
755
+ {
756
+ "epoch": 0.93,
757
+ "learning_rate": 3.4441125217824246e-05,
758
+ "loss": 1.6502,
759
+ "step": 1250
760
+ },
761
+ {
762
+ "epoch": 0.94,
763
+ "learning_rate": 3.431665421956684e-05,
764
+ "loss": 1.6617,
765
+ "step": 1260
766
+ },
767
+ {
768
+ "epoch": 0.95,
769
+ "learning_rate": 3.4192183221309436e-05,
770
+ "loss": 1.7155,
771
+ "step": 1270
772
+ },
773
+ {
774
+ "epoch": 0.96,
775
+ "learning_rate": 3.406771222305203e-05,
776
+ "loss": 1.691,
777
+ "step": 1280
778
+ },
779
+ {
780
+ "epoch": 0.96,
781
+ "learning_rate": 3.3943241224794625e-05,
782
+ "loss": 1.6601,
783
+ "step": 1290
784
+ },
785
+ {
786
+ "epoch": 0.97,
787
+ "learning_rate": 3.3818770226537214e-05,
788
+ "loss": 1.6607,
789
+ "step": 1300
790
+ },
791
+ {
792
+ "epoch": 0.98,
793
+ "learning_rate": 3.3694299228279815e-05,
794
+ "loss": 1.7053,
795
+ "step": 1310
796
+ },
797
+ {
798
+ "epoch": 0.99,
799
+ "learning_rate": 3.35698282300224e-05,
800
+ "loss": 1.6681,
801
+ "step": 1320
802
+ },
803
+ {
804
+ "epoch": 0.99,
805
+ "learning_rate": 3.3445357231765e-05,
806
+ "loss": 1.6712,
807
+ "step": 1330
808
+ },
809
+ {
810
+ "epoch": 1.0,
811
+ "learning_rate": 3.332088623350759e-05,
812
+ "loss": 1.6601,
813
+ "step": 1340
814
+ },
815
+ {
816
+ "epoch": 1.01,
817
+ "learning_rate": 3.319641523525019e-05,
818
+ "loss": 1.6522,
819
+ "step": 1350
820
+ },
821
+ {
822
+ "epoch": 1.02,
823
+ "learning_rate": 3.307194423699278e-05,
824
+ "loss": 1.6621,
825
+ "step": 1360
826
+ },
827
+ {
828
+ "epoch": 1.02,
829
+ "learning_rate": 3.294747323873538e-05,
830
+ "loss": 1.6651,
831
+ "step": 1370
832
+ },
833
+ {
834
+ "epoch": 1.03,
835
+ "learning_rate": 3.2823002240477966e-05,
836
+ "loss": 1.6665,
837
+ "step": 1380
838
+ },
839
+ {
840
+ "epoch": 1.04,
841
+ "learning_rate": 3.269853124222057e-05,
842
+ "loss": 1.6811,
843
+ "step": 1390
844
+ },
845
+ {
846
+ "epoch": 1.05,
847
+ "learning_rate": 3.2574060243963156e-05,
848
+ "loss": 1.6632,
849
+ "step": 1400
850
+ },
851
+ {
852
+ "epoch": 1.05,
853
+ "learning_rate": 3.244958924570575e-05,
854
+ "loss": 1.6529,
855
+ "step": 1410
856
+ },
857
+ {
858
+ "epoch": 1.06,
859
+ "learning_rate": 3.2325118247448345e-05,
860
+ "loss": 1.6943,
861
+ "step": 1420
862
+ },
863
+ {
864
+ "epoch": 1.07,
865
+ "learning_rate": 3.220064724919094e-05,
866
+ "loss": 1.6932,
867
+ "step": 1430
868
+ },
869
+ {
870
+ "epoch": 1.08,
871
+ "learning_rate": 3.2076176250933535e-05,
872
+ "loss": 1.6711,
873
+ "step": 1440
874
+ },
875
+ {
876
+ "epoch": 1.08,
877
+ "learning_rate": 3.195170525267613e-05,
878
+ "loss": 1.6711,
879
+ "step": 1450
880
+ },
881
+ {
882
+ "epoch": 1.09,
883
+ "learning_rate": 3.182723425441872e-05,
884
+ "loss": 1.6678,
885
+ "step": 1460
886
+ },
887
+ {
888
+ "epoch": 1.1,
889
+ "learning_rate": 3.170276325616132e-05,
890
+ "loss": 1.7058,
891
+ "step": 1470
892
+ },
893
+ {
894
+ "epoch": 1.11,
895
+ "learning_rate": 3.157829225790391e-05,
896
+ "loss": 1.7017,
897
+ "step": 1480
898
+ },
899
+ {
900
+ "epoch": 1.11,
901
+ "learning_rate": 3.14538212596465e-05,
902
+ "loss": 1.6862,
903
+ "step": 1490
904
+ },
905
+ {
906
+ "epoch": 1.12,
907
+ "learning_rate": 3.13293502613891e-05,
908
+ "loss": 1.6425,
909
+ "step": 1500
910
+ },
911
+ {
912
+ "epoch": 1.13,
913
+ "learning_rate": 3.120487926313169e-05,
914
+ "loss": 1.6977,
915
+ "step": 1510
916
+ },
917
+ {
918
+ "epoch": 1.14,
919
+ "learning_rate": 3.108040826487428e-05,
920
+ "loss": 1.698,
921
+ "step": 1520
922
+ },
923
+ {
924
+ "epoch": 1.14,
925
+ "learning_rate": 3.095593726661688e-05,
926
+ "loss": 1.6718,
927
+ "step": 1530
928
+ },
929
+ {
930
+ "epoch": 1.15,
931
+ "learning_rate": 3.083146626835947e-05,
932
+ "loss": 1.7025,
933
+ "step": 1540
934
+ },
935
+ {
936
+ "epoch": 1.16,
937
+ "learning_rate": 3.0706995270102065e-05,
938
+ "loss": 1.665,
939
+ "step": 1550
940
+ },
941
+ {
942
+ "epoch": 1.17,
943
+ "learning_rate": 3.058252427184466e-05,
944
+ "loss": 1.7035,
945
+ "step": 1560
946
+ },
947
+ {
948
+ "epoch": 1.17,
949
+ "learning_rate": 3.0458053273587255e-05,
950
+ "loss": 1.6849,
951
+ "step": 1570
952
+ },
953
+ {
954
+ "epoch": 1.18,
955
+ "learning_rate": 3.0333582275329846e-05,
956
+ "loss": 1.6345,
957
+ "step": 1580
958
+ },
959
+ {
960
+ "epoch": 1.19,
961
+ "learning_rate": 3.0209111277072444e-05,
962
+ "loss": 1.6593,
963
+ "step": 1590
964
+ },
965
+ {
966
+ "epoch": 1.19,
967
+ "learning_rate": 3.0084640278815036e-05,
968
+ "loss": 1.6248,
969
+ "step": 1600
970
+ },
971
+ {
972
+ "epoch": 1.2,
973
+ "learning_rate": 2.996016928055763e-05,
974
+ "loss": 1.6827,
975
+ "step": 1610
976
+ },
977
+ {
978
+ "epoch": 1.21,
979
+ "learning_rate": 2.9835698282300222e-05,
980
+ "loss": 1.7054,
981
+ "step": 1620
982
+ },
983
+ {
984
+ "epoch": 1.22,
985
+ "learning_rate": 2.971122728404282e-05,
986
+ "loss": 1.6518,
987
+ "step": 1630
988
+ },
989
+ {
990
+ "epoch": 1.22,
991
+ "learning_rate": 2.9586756285785412e-05,
992
+ "loss": 1.6795,
993
+ "step": 1640
994
+ },
995
+ {
996
+ "epoch": 1.23,
997
+ "learning_rate": 2.9462285287528007e-05,
998
+ "loss": 1.6286,
999
+ "step": 1650
1000
+ },
1001
+ {
1002
+ "epoch": 1.24,
1003
+ "learning_rate": 2.93378142892706e-05,
1004
+ "loss": 1.6599,
1005
+ "step": 1660
1006
+ },
1007
+ {
1008
+ "epoch": 1.25,
1009
+ "learning_rate": 2.9213343291013197e-05,
1010
+ "loss": 1.6902,
1011
+ "step": 1670
1012
+ },
1013
+ {
1014
+ "epoch": 1.25,
1015
+ "learning_rate": 2.9088872292755788e-05,
1016
+ "loss": 1.6551,
1017
+ "step": 1680
1018
+ },
1019
+ {
1020
+ "epoch": 1.26,
1021
+ "learning_rate": 2.8964401294498383e-05,
1022
+ "loss": 1.6493,
1023
+ "step": 1690
1024
+ },
1025
+ {
1026
+ "epoch": 1.27,
1027
+ "learning_rate": 2.8839930296240975e-05,
1028
+ "loss": 1.7094,
1029
+ "step": 1700
1030
+ },
1031
+ {
1032
+ "epoch": 1.28,
1033
+ "learning_rate": 2.8715459297983573e-05,
1034
+ "loss": 1.6714,
1035
+ "step": 1710
1036
+ },
1037
+ {
1038
+ "epoch": 1.28,
1039
+ "learning_rate": 2.859098829972616e-05,
1040
+ "loss": 1.6296,
1041
+ "step": 1720
1042
+ },
1043
+ {
1044
+ "epoch": 1.29,
1045
+ "learning_rate": 2.846651730146876e-05,
1046
+ "loss": 1.6882,
1047
+ "step": 1730
1048
+ },
1049
+ {
1050
+ "epoch": 1.3,
1051
+ "learning_rate": 2.834204630321135e-05,
1052
+ "loss": 1.6493,
1053
+ "step": 1740
1054
+ },
1055
+ {
1056
+ "epoch": 1.31,
1057
+ "learning_rate": 2.8217575304953945e-05,
1058
+ "loss": 1.6489,
1059
+ "step": 1750
1060
+ },
1061
+ {
1062
+ "epoch": 1.31,
1063
+ "learning_rate": 2.8093104306696544e-05,
1064
+ "loss": 1.6683,
1065
+ "step": 1760
1066
+ },
1067
+ {
1068
+ "epoch": 1.32,
1069
+ "learning_rate": 2.7968633308439135e-05,
1070
+ "loss": 1.6988,
1071
+ "step": 1770
1072
+ },
1073
+ {
1074
+ "epoch": 1.33,
1075
+ "learning_rate": 2.784416231018173e-05,
1076
+ "loss": 1.648,
1077
+ "step": 1780
1078
+ },
1079
+ {
1080
+ "epoch": 1.34,
1081
+ "learning_rate": 2.771969131192432e-05,
1082
+ "loss": 1.6655,
1083
+ "step": 1790
1084
+ },
1085
+ {
1086
+ "epoch": 1.34,
1087
+ "learning_rate": 2.759522031366692e-05,
1088
+ "loss": 1.6709,
1089
+ "step": 1800
1090
+ },
1091
+ {
1092
+ "epoch": 1.35,
1093
+ "learning_rate": 2.747074931540951e-05,
1094
+ "loss": 1.665,
1095
+ "step": 1810
1096
+ },
1097
+ {
1098
+ "epoch": 1.36,
1099
+ "learning_rate": 2.7346278317152106e-05,
1100
+ "loss": 1.6411,
1101
+ "step": 1820
1102
+ },
1103
+ {
1104
+ "epoch": 1.37,
1105
+ "learning_rate": 2.7221807318894698e-05,
1106
+ "loss": 1.6913,
1107
+ "step": 1830
1108
+ },
1109
+ {
1110
+ "epoch": 1.37,
1111
+ "learning_rate": 2.7097336320637296e-05,
1112
+ "loss": 1.6906,
1113
+ "step": 1840
1114
+ },
1115
+ {
1116
+ "epoch": 1.38,
1117
+ "learning_rate": 2.6972865322379887e-05,
1118
+ "loss": 1.6345,
1119
+ "step": 1850
1120
+ },
1121
+ {
1122
+ "epoch": 1.39,
1123
+ "learning_rate": 2.6848394324122482e-05,
1124
+ "loss": 1.6903,
1125
+ "step": 1860
1126
+ },
1127
+ {
1128
+ "epoch": 1.4,
1129
+ "learning_rate": 2.6723923325865074e-05,
1130
+ "loss": 1.6819,
1131
+ "step": 1870
1132
+ },
1133
+ {
1134
+ "epoch": 1.4,
1135
+ "learning_rate": 2.6599452327607672e-05,
1136
+ "loss": 1.6393,
1137
+ "step": 1880
1138
+ },
1139
+ {
1140
+ "epoch": 1.41,
1141
+ "learning_rate": 2.6474981329350264e-05,
1142
+ "loss": 1.6725,
1143
+ "step": 1890
1144
+ },
1145
+ {
1146
+ "epoch": 1.42,
1147
+ "learning_rate": 2.635051033109286e-05,
1148
+ "loss": 1.6736,
1149
+ "step": 1900
1150
+ },
1151
+ {
1152
+ "epoch": 1.43,
1153
+ "learning_rate": 2.622603933283545e-05,
1154
+ "loss": 1.6734,
1155
+ "step": 1910
1156
+ },
1157
+ {
1158
+ "epoch": 1.43,
1159
+ "learning_rate": 2.6101568334578048e-05,
1160
+ "loss": 1.6761,
1161
+ "step": 1920
1162
+ },
1163
+ {
1164
+ "epoch": 1.44,
1165
+ "learning_rate": 2.597709733632064e-05,
1166
+ "loss": 1.6842,
1167
+ "step": 1930
1168
+ },
1169
+ {
1170
+ "epoch": 1.45,
1171
+ "learning_rate": 2.5852626338063234e-05,
1172
+ "loss": 1.7064,
1173
+ "step": 1940
1174
+ },
1175
+ {
1176
+ "epoch": 1.46,
1177
+ "learning_rate": 2.5728155339805826e-05,
1178
+ "loss": 1.7036,
1179
+ "step": 1950
1180
+ },
1181
+ {
1182
+ "epoch": 1.46,
1183
+ "learning_rate": 2.5603684341548424e-05,
1184
+ "loss": 1.6563,
1185
+ "step": 1960
1186
+ },
1187
+ {
1188
+ "epoch": 1.47,
1189
+ "learning_rate": 2.5479213343291016e-05,
1190
+ "loss": 1.6796,
1191
+ "step": 1970
1192
+ },
1193
+ {
1194
+ "epoch": 1.48,
1195
+ "learning_rate": 2.535474234503361e-05,
1196
+ "loss": 1.6469,
1197
+ "step": 1980
1198
+ },
1199
+ {
1200
+ "epoch": 1.49,
1201
+ "learning_rate": 2.5230271346776202e-05,
1202
+ "loss": 1.6681,
1203
+ "step": 1990
1204
+ },
1205
+ {
1206
+ "epoch": 1.49,
1207
+ "learning_rate": 2.51058003485188e-05,
1208
+ "loss": 1.6632,
1209
+ "step": 2000
1210
+ },
1211
+ {
1212
+ "epoch": 1.5,
1213
+ "learning_rate": 2.498132935026139e-05,
1214
+ "loss": 1.6799,
1215
+ "step": 2010
1216
+ },
1217
+ {
1218
+ "epoch": 1.51,
1219
+ "learning_rate": 2.4856858352003983e-05,
1220
+ "loss": 1.6811,
1221
+ "step": 2020
1222
+ },
1223
+ {
1224
+ "epoch": 1.52,
1225
+ "learning_rate": 2.4732387353746578e-05,
1226
+ "loss": 1.6365,
1227
+ "step": 2030
1228
+ },
1229
+ {
1230
+ "epoch": 1.52,
1231
+ "learning_rate": 2.4607916355489173e-05,
1232
+ "loss": 1.6773,
1233
+ "step": 2040
1234
+ },
1235
+ {
1236
+ "epoch": 1.53,
1237
+ "learning_rate": 2.4483445357231764e-05,
1238
+ "loss": 1.6194,
1239
+ "step": 2050
1240
+ },
1241
+ {
1242
+ "epoch": 1.54,
1243
+ "learning_rate": 2.435897435897436e-05,
1244
+ "loss": 1.6865,
1245
+ "step": 2060
1246
+ },
1247
+ {
1248
+ "epoch": 1.55,
1249
+ "learning_rate": 2.4234503360716954e-05,
1250
+ "loss": 1.6552,
1251
+ "step": 2070
1252
+ },
1253
+ {
1254
+ "epoch": 1.55,
1255
+ "learning_rate": 2.411003236245955e-05,
1256
+ "loss": 1.6823,
1257
+ "step": 2080
1258
+ },
1259
+ {
1260
+ "epoch": 1.56,
1261
+ "learning_rate": 2.398556136420214e-05,
1262
+ "loss": 1.6611,
1263
+ "step": 2090
1264
+ },
1265
+ {
1266
+ "epoch": 1.57,
1267
+ "learning_rate": 2.3861090365944735e-05,
1268
+ "loss": 1.6662,
1269
+ "step": 2100
1270
+ },
1271
+ {
1272
+ "epoch": 1.58,
1273
+ "learning_rate": 2.373661936768733e-05,
1274
+ "loss": 1.6898,
1275
+ "step": 2110
1276
+ },
1277
+ {
1278
+ "epoch": 1.58,
1279
+ "learning_rate": 2.3612148369429922e-05,
1280
+ "loss": 1.6474,
1281
+ "step": 2120
1282
+ },
1283
+ {
1284
+ "epoch": 1.59,
1285
+ "learning_rate": 2.3487677371172517e-05,
1286
+ "loss": 1.6635,
1287
+ "step": 2130
1288
+ },
1289
+ {
1290
+ "epoch": 1.6,
1291
+ "learning_rate": 2.336320637291511e-05,
1292
+ "loss": 1.6274,
1293
+ "step": 2140
1294
+ },
1295
+ {
1296
+ "epoch": 1.61,
1297
+ "learning_rate": 2.3238735374657706e-05,
1298
+ "loss": 1.6794,
1299
+ "step": 2150
1300
+ },
1301
+ {
1302
+ "epoch": 1.61,
1303
+ "learning_rate": 2.3114264376400298e-05,
1304
+ "loss": 1.6411,
1305
+ "step": 2160
1306
+ },
1307
+ {
1308
+ "epoch": 1.62,
1309
+ "learning_rate": 2.2989793378142893e-05,
1310
+ "loss": 1.6793,
1311
+ "step": 2170
1312
+ },
1313
+ {
1314
+ "epoch": 1.63,
1315
+ "learning_rate": 2.2865322379885488e-05,
1316
+ "loss": 1.6772,
1317
+ "step": 2180
1318
+ },
1319
+ {
1320
+ "epoch": 1.64,
1321
+ "learning_rate": 2.2740851381628083e-05,
1322
+ "loss": 1.6425,
1323
+ "step": 2190
1324
+ },
1325
+ {
1326
+ "epoch": 1.64,
1327
+ "learning_rate": 2.2616380383370674e-05,
1328
+ "loss": 1.6614,
1329
+ "step": 2200
1330
+ },
1331
+ {
1332
+ "epoch": 1.65,
1333
+ "learning_rate": 2.249190938511327e-05,
1334
+ "loss": 1.65,
1335
+ "step": 2210
1336
+ },
1337
+ {
1338
+ "epoch": 1.66,
1339
+ "learning_rate": 2.2367438386855864e-05,
1340
+ "loss": 1.685,
1341
+ "step": 2220
1342
+ },
1343
+ {
1344
+ "epoch": 1.67,
1345
+ "learning_rate": 2.2242967388598455e-05,
1346
+ "loss": 1.6808,
1347
+ "step": 2230
1348
+ },
1349
+ {
1350
+ "epoch": 1.67,
1351
+ "learning_rate": 2.211849639034105e-05,
1352
+ "loss": 1.6612,
1353
+ "step": 2240
1354
+ },
1355
+ {
1356
+ "epoch": 1.68,
1357
+ "learning_rate": 2.1994025392083645e-05,
1358
+ "loss": 1.6631,
1359
+ "step": 2250
1360
+ },
1361
+ {
1362
+ "epoch": 1.69,
1363
+ "learning_rate": 2.186955439382624e-05,
1364
+ "loss": 1.6675,
1365
+ "step": 2260
1366
+ },
1367
+ {
1368
+ "epoch": 1.7,
1369
+ "learning_rate": 2.174508339556883e-05,
1370
+ "loss": 1.6532,
1371
+ "step": 2270
1372
+ },
1373
+ {
1374
+ "epoch": 1.7,
1375
+ "learning_rate": 2.1620612397311426e-05,
1376
+ "loss": 1.7217,
1377
+ "step": 2280
1378
+ },
1379
+ {
1380
+ "epoch": 1.71,
1381
+ "learning_rate": 2.149614139905402e-05,
1382
+ "loss": 1.6697,
1383
+ "step": 2290
1384
+ },
1385
+ {
1386
+ "epoch": 1.72,
1387
+ "learning_rate": 2.1371670400796616e-05,
1388
+ "loss": 1.634,
1389
+ "step": 2300
1390
+ },
1391
+ {
1392
+ "epoch": 1.73,
1393
+ "learning_rate": 2.1247199402539207e-05,
1394
+ "loss": 1.6968,
1395
+ "step": 2310
1396
+ },
1397
+ {
1398
+ "epoch": 1.73,
1399
+ "learning_rate": 2.1122728404281802e-05,
1400
+ "loss": 1.6925,
1401
+ "step": 2320
1402
+ },
1403
+ {
1404
+ "epoch": 1.74,
1405
+ "learning_rate": 2.0998257406024397e-05,
1406
+ "loss": 1.6455,
1407
+ "step": 2330
1408
+ },
1409
+ {
1410
+ "epoch": 1.75,
1411
+ "learning_rate": 2.0873786407766992e-05,
1412
+ "loss": 1.6365,
1413
+ "step": 2340
1414
+ },
1415
+ {
1416
+ "epoch": 1.76,
1417
+ "learning_rate": 2.0749315409509583e-05,
1418
+ "loss": 1.6813,
1419
+ "step": 2350
1420
+ },
1421
+ {
1422
+ "epoch": 1.76,
1423
+ "learning_rate": 2.062484441125218e-05,
1424
+ "loss": 1.6725,
1425
+ "step": 2360
1426
+ },
1427
+ {
1428
+ "epoch": 1.77,
1429
+ "learning_rate": 2.0500373412994773e-05,
1430
+ "loss": 1.6718,
1431
+ "step": 2370
1432
+ },
1433
+ {
1434
+ "epoch": 1.78,
1435
+ "learning_rate": 2.0375902414737365e-05,
1436
+ "loss": 1.6954,
1437
+ "step": 2380
1438
+ },
1439
+ {
1440
+ "epoch": 1.78,
1441
+ "learning_rate": 2.025143141647996e-05,
1442
+ "loss": 1.6719,
1443
+ "step": 2390
1444
+ },
1445
+ {
1446
+ "epoch": 1.79,
1447
+ "learning_rate": 2.0126960418222554e-05,
1448
+ "loss": 1.6549,
1449
+ "step": 2400
1450
+ },
1451
+ {
1452
+ "epoch": 1.8,
1453
+ "learning_rate": 2.000248941996515e-05,
1454
+ "loss": 1.6761,
1455
+ "step": 2410
1456
+ },
1457
+ {
1458
+ "epoch": 1.81,
1459
+ "learning_rate": 1.987801842170774e-05,
1460
+ "loss": 1.6582,
1461
+ "step": 2420
1462
+ },
1463
+ {
1464
+ "epoch": 1.81,
1465
+ "learning_rate": 1.9753547423450336e-05,
1466
+ "loss": 1.6202,
1467
+ "step": 2430
1468
+ },
1469
+ {
1470
+ "epoch": 1.82,
1471
+ "learning_rate": 1.962907642519293e-05,
1472
+ "loss": 1.6645,
1473
+ "step": 2440
1474
+ },
1475
+ {
1476
+ "epoch": 1.83,
1477
+ "learning_rate": 1.9504605426935525e-05,
1478
+ "loss": 1.6565,
1479
+ "step": 2450
1480
+ },
1481
+ {
1482
+ "epoch": 1.84,
1483
+ "learning_rate": 1.9380134428678117e-05,
1484
+ "loss": 1.6395,
1485
+ "step": 2460
1486
+ },
1487
+ {
1488
+ "epoch": 1.84,
1489
+ "learning_rate": 1.9255663430420712e-05,
1490
+ "loss": 1.6469,
1491
+ "step": 2470
1492
+ },
1493
+ {
1494
+ "epoch": 1.85,
1495
+ "learning_rate": 1.9131192432163307e-05,
1496
+ "loss": 1.6523,
1497
+ "step": 2480
1498
+ },
1499
+ {
1500
+ "epoch": 1.86,
1501
+ "learning_rate": 1.9006721433905898e-05,
1502
+ "loss": 1.6799,
1503
+ "step": 2490
1504
+ },
1505
+ {
1506
+ "epoch": 1.87,
1507
+ "learning_rate": 1.8882250435648493e-05,
1508
+ "loss": 1.6584,
1509
+ "step": 2500
1510
+ },
1511
+ {
1512
+ "epoch": 1.87,
1513
+ "learning_rate": 1.8757779437391088e-05,
1514
+ "loss": 1.6729,
1515
+ "step": 2510
1516
+ },
1517
+ {
1518
+ "epoch": 1.88,
1519
+ "learning_rate": 1.8633308439133683e-05,
1520
+ "loss": 1.6157,
1521
+ "step": 2520
1522
+ },
1523
+ {
1524
+ "epoch": 1.89,
1525
+ "learning_rate": 1.8508837440876274e-05,
1526
+ "loss": 1.7094,
1527
+ "step": 2530
1528
+ },
1529
+ {
1530
+ "epoch": 1.9,
1531
+ "learning_rate": 1.838436644261887e-05,
1532
+ "loss": 1.68,
1533
+ "step": 2540
1534
+ },
1535
+ {
1536
+ "epoch": 1.9,
1537
+ "learning_rate": 1.8259895444361464e-05,
1538
+ "loss": 1.7009,
1539
+ "step": 2550
1540
+ },
1541
+ {
1542
+ "epoch": 1.91,
1543
+ "learning_rate": 1.813542444610406e-05,
1544
+ "loss": 1.6931,
1545
+ "step": 2560
1546
+ },
1547
+ {
1548
+ "epoch": 1.92,
1549
+ "learning_rate": 1.801095344784665e-05,
1550
+ "loss": 1.7001,
1551
+ "step": 2570
1552
+ },
1553
+ {
1554
+ "epoch": 1.93,
1555
+ "learning_rate": 1.7886482449589245e-05,
1556
+ "loss": 1.6847,
1557
+ "step": 2580
1558
+ },
1559
+ {
1560
+ "epoch": 1.93,
1561
+ "learning_rate": 1.776201145133184e-05,
1562
+ "loss": 1.653,
1563
+ "step": 2590
1564
+ },
1565
+ {
1566
+ "epoch": 1.94,
1567
+ "learning_rate": 1.763754045307443e-05,
1568
+ "loss": 1.6386,
1569
+ "step": 2600
1570
+ },
1571
+ {
1572
+ "epoch": 1.95,
1573
+ "learning_rate": 1.7513069454817026e-05,
1574
+ "loss": 1.6832,
1575
+ "step": 2610
1576
+ },
1577
+ {
1578
+ "epoch": 1.96,
1579
+ "learning_rate": 1.7388598456559625e-05,
1580
+ "loss": 1.6853,
1581
+ "step": 2620
1582
+ },
1583
+ {
1584
+ "epoch": 1.96,
1585
+ "learning_rate": 1.7264127458302216e-05,
1586
+ "loss": 1.6876,
1587
+ "step": 2630
1588
+ },
1589
+ {
1590
+ "epoch": 1.97,
1591
+ "learning_rate": 1.713965646004481e-05,
1592
+ "loss": 1.6878,
1593
+ "step": 2640
1594
+ },
1595
+ {
1596
+ "epoch": 1.98,
1597
+ "learning_rate": 1.7015185461787406e-05,
1598
+ "loss": 1.7067,
1599
+ "step": 2650
1600
+ },
1601
+ {
1602
+ "epoch": 1.99,
1603
+ "learning_rate": 1.689071446353e-05,
1604
+ "loss": 1.6736,
1605
+ "step": 2660
1606
+ },
1607
+ {
1608
+ "epoch": 1.99,
1609
+ "learning_rate": 1.6766243465272592e-05,
1610
+ "loss": 1.6536,
1611
+ "step": 2670
1612
+ },
1613
+ {
1614
+ "epoch": 2.0,
1615
+ "learning_rate": 1.6641772467015187e-05,
1616
+ "loss": 1.6806,
1617
+ "step": 2680
1618
+ },
1619
+ {
1620
+ "epoch": 2.01,
1621
+ "learning_rate": 1.6517301468757782e-05,
1622
+ "loss": 1.6708,
1623
+ "step": 2690
1624
+ },
1625
+ {
1626
+ "epoch": 2.02,
1627
+ "learning_rate": 1.6392830470500377e-05,
1628
+ "loss": 1.6718,
1629
+ "step": 2700
1630
+ },
1631
+ {
1632
+ "epoch": 2.02,
1633
+ "learning_rate": 1.626835947224297e-05,
1634
+ "loss": 1.6829,
1635
+ "step": 2710
1636
+ },
1637
+ {
1638
+ "epoch": 2.03,
1639
+ "learning_rate": 1.6143888473985563e-05,
1640
+ "loss": 1.7071,
1641
+ "step": 2720
1642
+ },
1643
+ {
1644
+ "epoch": 2.04,
1645
+ "learning_rate": 1.6019417475728158e-05,
1646
+ "loss": 1.6583,
1647
+ "step": 2730
1648
+ },
1649
+ {
1650
+ "epoch": 2.05,
1651
+ "learning_rate": 1.5894946477470753e-05,
1652
+ "loss": 1.6634,
1653
+ "step": 2740
1654
+ },
1655
+ {
1656
+ "epoch": 2.05,
1657
+ "learning_rate": 1.5770475479213344e-05,
1658
+ "loss": 1.6614,
1659
+ "step": 2750
1660
+ },
1661
+ {
1662
+ "epoch": 2.06,
1663
+ "learning_rate": 1.564600448095594e-05,
1664
+ "loss": 1.6689,
1665
+ "step": 2760
1666
+ },
1667
+ {
1668
+ "epoch": 2.07,
1669
+ "learning_rate": 1.5521533482698534e-05,
1670
+ "loss": 1.6578,
1671
+ "step": 2770
1672
+ },
1673
+ {
1674
+ "epoch": 2.08,
1675
+ "learning_rate": 1.5397062484441126e-05,
1676
+ "loss": 1.6791,
1677
+ "step": 2780
1678
+ },
1679
+ {
1680
+ "epoch": 2.08,
1681
+ "learning_rate": 1.527259148618372e-05,
1682
+ "loss": 1.6402,
1683
+ "step": 2790
1684
+ },
1685
+ {
1686
+ "epoch": 2.09,
1687
+ "learning_rate": 1.5148120487926315e-05,
1688
+ "loss": 1.6574,
1689
+ "step": 2800
1690
+ },
1691
+ {
1692
+ "epoch": 2.1,
1693
+ "learning_rate": 1.5023649489668909e-05,
1694
+ "loss": 1.6598,
1695
+ "step": 2810
1696
+ },
1697
+ {
1698
+ "epoch": 2.11,
1699
+ "learning_rate": 1.4899178491411503e-05,
1700
+ "loss": 1.6827,
1701
+ "step": 2820
1702
+ },
1703
+ {
1704
+ "epoch": 2.11,
1705
+ "learning_rate": 1.4774707493154097e-05,
1706
+ "loss": 1.6471,
1707
+ "step": 2830
1708
+ },
1709
+ {
1710
+ "epoch": 2.12,
1711
+ "learning_rate": 1.4650236494896691e-05,
1712
+ "loss": 1.6398,
1713
+ "step": 2840
1714
+ },
1715
+ {
1716
+ "epoch": 2.13,
1717
+ "learning_rate": 1.4525765496639285e-05,
1718
+ "loss": 1.6557,
1719
+ "step": 2850
1720
+ },
1721
+ {
1722
+ "epoch": 2.14,
1723
+ "learning_rate": 1.440129449838188e-05,
1724
+ "loss": 1.6614,
1725
+ "step": 2860
1726
+ },
1727
+ {
1728
+ "epoch": 2.14,
1729
+ "learning_rate": 1.4276823500124473e-05,
1730
+ "loss": 1.6619,
1731
+ "step": 2870
1732
+ },
1733
+ {
1734
+ "epoch": 2.15,
1735
+ "learning_rate": 1.4152352501867066e-05,
1736
+ "loss": 1.6887,
1737
+ "step": 2880
1738
+ },
1739
+ {
1740
+ "epoch": 2.16,
1741
+ "learning_rate": 1.402788150360966e-05,
1742
+ "loss": 1.6815,
1743
+ "step": 2890
1744
+ },
1745
+ {
1746
+ "epoch": 2.17,
1747
+ "learning_rate": 1.3903410505352254e-05,
1748
+ "loss": 1.6267,
1749
+ "step": 2900
1750
+ },
1751
+ {
1752
+ "epoch": 2.17,
1753
+ "learning_rate": 1.3778939507094849e-05,
1754
+ "loss": 1.6938,
1755
+ "step": 2910
1756
+ },
1757
+ {
1758
+ "epoch": 2.18,
1759
+ "learning_rate": 1.3654468508837442e-05,
1760
+ "loss": 1.6796,
1761
+ "step": 2920
1762
+ },
1763
+ {
1764
+ "epoch": 2.19,
1765
+ "learning_rate": 1.3529997510580037e-05,
1766
+ "loss": 1.6544,
1767
+ "step": 2930
1768
+ },
1769
+ {
1770
+ "epoch": 2.2,
1771
+ "learning_rate": 1.340552651232263e-05,
1772
+ "loss": 1.6497,
1773
+ "step": 2940
1774
+ },
1775
+ {
1776
+ "epoch": 2.2,
1777
+ "learning_rate": 1.3281055514065225e-05,
1778
+ "loss": 1.6397,
1779
+ "step": 2950
1780
+ },
1781
+ {
1782
+ "epoch": 2.21,
1783
+ "learning_rate": 1.3156584515807818e-05,
1784
+ "loss": 1.6788,
1785
+ "step": 2960
1786
+ },
1787
+ {
1788
+ "epoch": 2.22,
1789
+ "learning_rate": 1.3032113517550413e-05,
1790
+ "loss": 1.6675,
1791
+ "step": 2970
1792
+ },
1793
+ {
1794
+ "epoch": 2.23,
1795
+ "learning_rate": 1.2907642519293006e-05,
1796
+ "loss": 1.6429,
1797
+ "step": 2980
1798
+ },
1799
+ {
1800
+ "epoch": 2.23,
1801
+ "learning_rate": 1.27831715210356e-05,
1802
+ "loss": 1.6866,
1803
+ "step": 2990
1804
+ },
1805
+ {
1806
+ "epoch": 2.24,
1807
+ "learning_rate": 1.2658700522778194e-05,
1808
+ "loss": 1.6773,
1809
+ "step": 3000
1810
+ },
1811
+ {
1812
+ "epoch": 2.25,
1813
+ "learning_rate": 1.2534229524520787e-05,
1814
+ "loss": 1.6776,
1815
+ "step": 3010
1816
+ },
1817
+ {
1818
+ "epoch": 2.26,
1819
+ "learning_rate": 1.240975852626338e-05,
1820
+ "loss": 1.6598,
1821
+ "step": 3020
1822
+ },
1823
+ {
1824
+ "epoch": 2.26,
1825
+ "learning_rate": 1.2285287528005974e-05,
1826
+ "loss": 1.6904,
1827
+ "step": 3030
1828
+ },
1829
+ {
1830
+ "epoch": 2.27,
1831
+ "learning_rate": 1.2160816529748569e-05,
1832
+ "loss": 1.6412,
1833
+ "step": 3040
1834
+ },
1835
+ {
1836
+ "epoch": 2.28,
1837
+ "learning_rate": 1.2036345531491162e-05,
1838
+ "loss": 1.6352,
1839
+ "step": 3050
1840
+ },
1841
+ {
1842
+ "epoch": 2.29,
1843
+ "learning_rate": 1.1911874533233758e-05,
1844
+ "loss": 1.6672,
1845
+ "step": 3060
1846
+ },
1847
+ {
1848
+ "epoch": 2.29,
1849
+ "learning_rate": 1.1787403534976351e-05,
1850
+ "loss": 1.6616,
1851
+ "step": 3070
1852
+ },
1853
+ {
1854
+ "epoch": 2.3,
1855
+ "learning_rate": 1.1662932536718946e-05,
1856
+ "loss": 1.7068,
1857
+ "step": 3080
1858
+ },
1859
+ {
1860
+ "epoch": 2.31,
1861
+ "learning_rate": 1.153846153846154e-05,
1862
+ "loss": 1.6708,
1863
+ "step": 3090
1864
+ },
1865
+ {
1866
+ "epoch": 2.32,
1867
+ "learning_rate": 1.1413990540204134e-05,
1868
+ "loss": 1.6686,
1869
+ "step": 3100
1870
+ },
1871
+ {
1872
+ "epoch": 2.32,
1873
+ "learning_rate": 1.1289519541946728e-05,
1874
+ "loss": 1.6841,
1875
+ "step": 3110
1876
+ },
1877
+ {
1878
+ "epoch": 2.33,
1879
+ "learning_rate": 1.116504854368932e-05,
1880
+ "loss": 1.6632,
1881
+ "step": 3120
1882
+ },
1883
+ {
1884
+ "epoch": 2.34,
1885
+ "learning_rate": 1.1040577545431916e-05,
1886
+ "loss": 1.6638,
1887
+ "step": 3130
1888
+ },
1889
+ {
1890
+ "epoch": 2.35,
1891
+ "learning_rate": 1.0916106547174509e-05,
1892
+ "loss": 1.6495,
1893
+ "step": 3140
1894
+ },
1895
+ {
1896
+ "epoch": 2.35,
1897
+ "learning_rate": 1.0791635548917104e-05,
1898
+ "loss": 1.6568,
1899
+ "step": 3150
1900
+ },
1901
+ {
1902
+ "epoch": 2.36,
1903
+ "learning_rate": 1.0667164550659697e-05,
1904
+ "loss": 1.6986,
1905
+ "step": 3160
1906
+ },
1907
+ {
1908
+ "epoch": 2.37,
1909
+ "learning_rate": 1.0542693552402292e-05,
1910
+ "loss": 1.6572,
1911
+ "step": 3170
1912
+ },
1913
+ {
1914
+ "epoch": 2.37,
1915
+ "learning_rate": 1.0418222554144885e-05,
1916
+ "loss": 1.6909,
1917
+ "step": 3180
1918
+ },
1919
+ {
1920
+ "epoch": 2.38,
1921
+ "learning_rate": 1.029375155588748e-05,
1922
+ "loss": 1.6765,
1923
+ "step": 3190
1924
+ },
1925
+ {
1926
+ "epoch": 2.39,
1927
+ "learning_rate": 1.0169280557630073e-05,
1928
+ "loss": 1.6493,
1929
+ "step": 3200
1930
+ },
1931
+ {
1932
+ "epoch": 2.4,
1933
+ "learning_rate": 1.0044809559372668e-05,
1934
+ "loss": 1.6784,
1935
+ "step": 3210
1936
+ },
1937
+ {
1938
+ "epoch": 2.4,
1939
+ "learning_rate": 9.920338561115261e-06,
1940
+ "loss": 1.6422,
1941
+ "step": 3220
1942
+ },
1943
+ {
1944
+ "epoch": 2.41,
1945
+ "learning_rate": 9.795867562857854e-06,
1946
+ "loss": 1.6546,
1947
+ "step": 3230
1948
+ },
1949
+ {
1950
+ "epoch": 2.42,
1951
+ "learning_rate": 9.671396564600449e-06,
1952
+ "loss": 1.6942,
1953
+ "step": 3240
1954
+ },
1955
+ {
1956
+ "epoch": 2.43,
1957
+ "learning_rate": 9.546925566343042e-06,
1958
+ "loss": 1.6706,
1959
+ "step": 3250
1960
+ },
1961
+ {
1962
+ "epoch": 2.43,
1963
+ "learning_rate": 9.422454568085637e-06,
1964
+ "loss": 1.6447,
1965
+ "step": 3260
1966
+ },
1967
+ {
1968
+ "epoch": 2.44,
1969
+ "learning_rate": 9.29798356982823e-06,
1970
+ "loss": 1.6744,
1971
+ "step": 3270
1972
+ },
1973
+ {
1974
+ "epoch": 2.45,
1975
+ "learning_rate": 9.173512571570825e-06,
1976
+ "loss": 1.6681,
1977
+ "step": 3280
1978
+ },
1979
+ {
1980
+ "epoch": 2.46,
1981
+ "learning_rate": 9.049041573313418e-06,
1982
+ "loss": 1.6916,
1983
+ "step": 3290
1984
+ },
1985
+ {
1986
+ "epoch": 2.46,
1987
+ "learning_rate": 8.924570575056013e-06,
1988
+ "loss": 1.7074,
1989
+ "step": 3300
1990
+ },
1991
+ {
1992
+ "epoch": 2.47,
1993
+ "learning_rate": 8.800099576798606e-06,
1994
+ "loss": 1.6726,
1995
+ "step": 3310
1996
+ },
1997
+ {
1998
+ "epoch": 2.48,
1999
+ "learning_rate": 8.675628578541201e-06,
2000
+ "loss": 1.6286,
2001
+ "step": 3320
2002
+ },
2003
+ {
2004
+ "epoch": 2.49,
2005
+ "learning_rate": 8.551157580283794e-06,
2006
+ "loss": 1.6319,
2007
+ "step": 3330
2008
+ },
2009
+ {
2010
+ "epoch": 2.49,
2011
+ "learning_rate": 8.42668658202639e-06,
2012
+ "loss": 1.6699,
2013
+ "step": 3340
2014
+ },
2015
+ {
2016
+ "epoch": 2.5,
2017
+ "learning_rate": 8.302215583768982e-06,
2018
+ "loss": 1.6123,
2019
+ "step": 3350
2020
+ },
2021
+ {
2022
+ "epoch": 2.51,
2023
+ "learning_rate": 8.177744585511576e-06,
2024
+ "loss": 1.6728,
2025
+ "step": 3360
2026
+ },
2027
+ {
2028
+ "epoch": 2.52,
2029
+ "learning_rate": 8.05327358725417e-06,
2030
+ "loss": 1.6663,
2031
+ "step": 3370
2032
+ },
2033
+ {
2034
+ "epoch": 2.52,
2035
+ "learning_rate": 7.928802588996764e-06,
2036
+ "loss": 1.6525,
2037
+ "step": 3380
2038
+ },
2039
+ {
2040
+ "epoch": 2.53,
2041
+ "learning_rate": 7.804331590739359e-06,
2042
+ "loss": 1.6713,
2043
+ "step": 3390
2044
+ },
2045
+ {
2046
+ "epoch": 2.54,
2047
+ "learning_rate": 7.679860592481952e-06,
2048
+ "loss": 1.635,
2049
+ "step": 3400
2050
+ },
2051
+ {
2052
+ "epoch": 2.55,
2053
+ "learning_rate": 7.555389594224547e-06,
2054
+ "loss": 1.6888,
2055
+ "step": 3410
2056
+ },
2057
+ {
2058
+ "epoch": 2.55,
2059
+ "learning_rate": 7.43091859596714e-06,
2060
+ "loss": 1.6957,
2061
+ "step": 3420
2062
+ },
2063
+ {
2064
+ "epoch": 2.56,
2065
+ "learning_rate": 7.306447597709734e-06,
2066
+ "loss": 1.6456,
2067
+ "step": 3430
2068
+ },
2069
+ {
2070
+ "epoch": 2.57,
2071
+ "learning_rate": 7.181976599452328e-06,
2072
+ "loss": 1.6401,
2073
+ "step": 3440
2074
+ },
2075
+ {
2076
+ "epoch": 2.58,
2077
+ "learning_rate": 7.057505601194922e-06,
2078
+ "loss": 1.6371,
2079
+ "step": 3450
2080
+ },
2081
+ {
2082
+ "epoch": 2.58,
2083
+ "learning_rate": 6.933034602937516e-06,
2084
+ "loss": 1.6487,
2085
+ "step": 3460
2086
+ },
2087
+ {
2088
+ "epoch": 2.59,
2089
+ "learning_rate": 6.80856360468011e-06,
2090
+ "loss": 1.6332,
2091
+ "step": 3470
2092
+ },
2093
+ {
2094
+ "epoch": 2.6,
2095
+ "learning_rate": 6.684092606422704e-06,
2096
+ "loss": 1.6792,
2097
+ "step": 3480
2098
+ },
2099
+ {
2100
+ "epoch": 2.61,
2101
+ "learning_rate": 6.559621608165298e-06,
2102
+ "loss": 1.6694,
2103
+ "step": 3490
2104
+ },
2105
+ {
2106
+ "epoch": 2.61,
2107
+ "learning_rate": 6.435150609907892e-06,
2108
+ "loss": 1.6753,
2109
+ "step": 3500
2110
+ }
2111
+ ],
2112
+ "logging_steps": 10,
2113
+ "max_steps": 4017,
2114
+ "num_input_tokens_seen": 0,
2115
+ "num_train_epochs": 3,
2116
+ "save_steps": 500,
2117
+ "total_flos": 2.21208687673344e+18,
2118
+ "train_batch_size": 4,
2119
+ "trial_name": null,
2120
+ "trial_params": null
2121
+ }