Model save
Browse files- README.md +4 -8
- all_results.json +5 -10
- config.json +1 -1
- model-00001-of-00003.safetensors +1 -1
- model-00002-of-00003.safetensors +1 -1
- model-00003-of-00003.safetensors +1 -1
- runs/Aug13_13-34-23_COE-CS-sv004/events.out.tfevents.1723556233.COE-CS-sv004.2672049.0 +3 -0
- train_results.json +5 -5
- trainer_state.json +206 -199
- training_args.bin +1 -1
README.md
CHANGED
@@ -2,15 +2,11 @@
|
|
2 |
license: apache-2.0
|
3 |
base_model: mistralai/Mistral-7B-Instruct-v0.2
|
4 |
tags:
|
5 |
-
- alignment-handbook
|
6 |
-
- trl
|
7 |
-
- sft
|
8 |
-
- generated_from_trainer
|
9 |
- trl
|
10 |
- sft
|
11 |
- generated_from_trainer
|
12 |
datasets:
|
13 |
-
-
|
14 |
model-index:
|
15 |
- name: preference_0.1p_seed42_level3_rare
|
16 |
results: []
|
@@ -21,9 +17,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
21 |
|
22 |
# preference_0.1p_seed42_level3_rare
|
23 |
|
24 |
-
This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) on the
|
25 |
It achieves the following results on the evaluation set:
|
26 |
-
- Loss: 0.
|
27 |
|
28 |
## Model description
|
29 |
|
@@ -60,7 +56,7 @@ The following hyperparameters were used during training:
|
|
60 |
|
61 |
| Training Loss | Epoch | Step | Validation Loss |
|
62 |
|:-------------:|:------:|:----:|:---------------:|
|
63 |
-
| 0.
|
64 |
|
65 |
|
66 |
### Framework versions
|
|
|
2 |
license: apache-2.0
|
3 |
base_model: mistralai/Mistral-7B-Instruct-v0.2
|
4 |
tags:
|
|
|
|
|
|
|
|
|
5 |
- trl
|
6 |
- sft
|
7 |
- generated_from_trainer
|
8 |
datasets:
|
9 |
+
- generator
|
10 |
model-index:
|
11 |
- name: preference_0.1p_seed42_level3_rare
|
12 |
results: []
|
|
|
17 |
|
18 |
# preference_0.1p_seed42_level3_rare
|
19 |
|
20 |
+
This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) on the generator dataset.
|
21 |
It achieves the following results on the evaluation set:
|
22 |
+
- Loss: 0.7390
|
23 |
|
24 |
## Model description
|
25 |
|
|
|
56 |
|
57 |
| Training Loss | Epoch | Step | Validation Loss |
|
58 |
|:-------------:|:------:|:----:|:---------------:|
|
59 |
+
| 0.6207 | 0.9979 | 234 | 0.7390 |
|
60 |
|
61 |
|
62 |
### Framework versions
|
all_results.json
CHANGED
@@ -1,14 +1,9 @@
|
|
1 |
{
|
2 |
-
"epoch": 0.
|
3 |
-
"
|
4 |
-
"
|
5 |
-
"
|
6 |
-
"eval_samples_per_second": 3.664,
|
7 |
-
"eval_steps_per_second": 0.921,
|
8 |
-
"total_flos": 47895596236800.0,
|
9 |
-
"train_loss": 0.7106172507506792,
|
10 |
-
"train_runtime": 7254.3633,
|
11 |
"train_samples": 18788,
|
12 |
-
"train_samples_per_second": 1.
|
13 |
"train_steps_per_second": 0.032
|
14 |
}
|
|
|
1 |
{
|
2 |
+
"epoch": 0.997867803837953,
|
3 |
+
"total_flos": 48942494515200.0,
|
4 |
+
"train_loss": 0.7695284368645432,
|
5 |
+
"train_runtime": 7306.9109,
|
|
|
|
|
|
|
|
|
|
|
6 |
"train_samples": 18788,
|
7 |
+
"train_samples_per_second": 1.026,
|
8 |
"train_steps_per_second": 0.032
|
9 |
}
|
config.json
CHANGED
@@ -22,6 +22,6 @@
|
|
22 |
"tie_word_embeddings": false,
|
23 |
"torch_dtype": "bfloat16",
|
24 |
"transformers_version": "4.43.4",
|
25 |
-
"use_cache":
|
26 |
"vocab_size": 32000
|
27 |
}
|
|
|
22 |
"tie_word_embeddings": false,
|
23 |
"torch_dtype": "bfloat16",
|
24 |
"transformers_version": "4.43.4",
|
25 |
+
"use_cache": false,
|
26 |
"vocab_size": 32000
|
27 |
}
|
model-00001-of-00003.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 4943162336
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:86f4d7f7f71205d776a15d9bb271ef8e4c5b85a4464a7d9cc8fabf5abbe6742a
|
3 |
size 4943162336
|
model-00002-of-00003.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 4999819336
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8d3da52201a469e680648b0686698f7ed8cc2c1f0cf1fb25f271d2e05cdf9b69
|
3 |
size 4999819336
|
model-00003-of-00003.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 4540516344
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e053a2e5d6ca1464762eb049d3dfabc9f0e6457f3bc4384424e2868e4071f4ba
|
3 |
size 4540516344
|
runs/Aug13_13-34-23_COE-CS-sv004/events.out.tfevents.1723556233.COE-CS-sv004.2672049.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:af08c6c58d2a3c6ef9cf9bc2649171c3215d1f8b4368f5fd3012ccbca822bf94
|
3 |
+
size 15669
|
train_results.json
CHANGED
@@ -1,9 +1,9 @@
|
|
1 |
{
|
2 |
-
"epoch": 0.
|
3 |
-
"total_flos":
|
4 |
-
"train_loss": 0.
|
5 |
-
"train_runtime":
|
6 |
"train_samples": 18788,
|
7 |
-
"train_samples_per_second": 1.
|
8 |
"train_steps_per_second": 0.032
|
9 |
}
|
|
|
1 |
{
|
2 |
+
"epoch": 0.997867803837953,
|
3 |
+
"total_flos": 48942494515200.0,
|
4 |
+
"train_loss": 0.7695284368645432,
|
5 |
+
"train_runtime": 7306.9109,
|
6 |
"train_samples": 18788,
|
7 |
+
"train_samples_per_second": 1.026,
|
8 |
"train_steps_per_second": 0.032
|
9 |
}
|
trainer_state.json
CHANGED
@@ -1,355 +1,362 @@
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
-
"epoch": 0.
|
5 |
"eval_steps": 500,
|
6 |
-
"global_step":
|
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.
|
13 |
-
"grad_norm":
|
14 |
-
"learning_rate": 4.
|
15 |
-
"loss": 1.
|
16 |
"step": 1
|
17 |
},
|
18 |
{
|
19 |
-
"epoch": 0.
|
20 |
-
"grad_norm":
|
21 |
-
"learning_rate": 2.
|
22 |
-
"loss": 1.
|
23 |
"step": 5
|
24 |
},
|
25 |
{
|
26 |
-
"epoch": 0.
|
27 |
-
"grad_norm": 3.
|
28 |
-
"learning_rate": 4.
|
29 |
-
"loss": 1.
|
30 |
"step": 10
|
31 |
},
|
32 |
{
|
33 |
-
"epoch": 0.
|
34 |
-
"grad_norm":
|
35 |
-
"learning_rate": 6.
|
36 |
-
"loss":
|
37 |
"step": 15
|
38 |
},
|
39 |
{
|
40 |
-
"epoch": 0.
|
41 |
-
"grad_norm": 2.
|
42 |
-
"learning_rate": 8.
|
43 |
-
"loss": 0.
|
44 |
"step": 20
|
45 |
},
|
46 |
{
|
47 |
-
"epoch": 0.
|
48 |
-
"grad_norm": 2.
|
49 |
-
"learning_rate": 9.
|
50 |
-
"loss": 0.
|
51 |
"step": 25
|
52 |
},
|
53 |
{
|
54 |
-
"epoch": 0.
|
55 |
-
"grad_norm": 2.
|
56 |
-
"learning_rate": 9.
|
57 |
-
"loss": 0.
|
58 |
"step": 30
|
59 |
},
|
60 |
{
|
61 |
-
"epoch": 0.
|
62 |
-
"grad_norm": 2.
|
63 |
-
"learning_rate": 9.
|
64 |
-
"loss": 0.
|
65 |
"step": 35
|
66 |
},
|
67 |
{
|
68 |
-
"epoch": 0.
|
69 |
-
"grad_norm": 2.
|
70 |
-
"learning_rate": 9.
|
71 |
-
"loss": 0.
|
72 |
"step": 40
|
73 |
},
|
74 |
{
|
75 |
-
"epoch": 0.
|
76 |
-
"grad_norm": 2.
|
77 |
-
"learning_rate": 9.
|
78 |
-
"loss": 0.
|
79 |
"step": 45
|
80 |
},
|
81 |
{
|
82 |
-
"epoch": 0.
|
83 |
-
"grad_norm": 2.
|
84 |
-
"learning_rate": 9.
|
85 |
-
"loss": 0.
|
86 |
"step": 50
|
87 |
},
|
88 |
{
|
89 |
-
"epoch": 0.
|
90 |
-
"grad_norm": 2.
|
91 |
-
"learning_rate": 9.
|
92 |
-
"loss": 0.
|
93 |
"step": 55
|
94 |
},
|
95 |
{
|
96 |
-
"epoch": 0.
|
97 |
-
"grad_norm": 2.
|
98 |
-
"learning_rate": 9.
|
99 |
-
"loss": 0.
|
100 |
"step": 60
|
101 |
},
|
102 |
{
|
103 |
-
"epoch": 0.
|
104 |
-
"grad_norm": 2.
|
105 |
-
"learning_rate": 9.
|
106 |
-
"loss": 0.
|
107 |
"step": 65
|
108 |
},
|
109 |
{
|
110 |
-
"epoch": 0.
|
111 |
-
"grad_norm":
|
112 |
-
"learning_rate": 8.
|
113 |
-
"loss": 0.
|
114 |
"step": 70
|
115 |
},
|
116 |
{
|
117 |
-
"epoch": 0.
|
118 |
-
"grad_norm": 2.
|
119 |
-
"learning_rate": 8.
|
120 |
-
"loss": 0.
|
121 |
"step": 75
|
122 |
},
|
123 |
{
|
124 |
-
"epoch": 0.
|
125 |
-
"grad_norm": 2.
|
126 |
-
"learning_rate": 8.
|
127 |
-
"loss": 0.
|
128 |
"step": 80
|
129 |
},
|
130 |
{
|
131 |
-
"epoch": 0.
|
132 |
-
"grad_norm":
|
133 |
-
"learning_rate":
|
134 |
-
"loss": 0.
|
135 |
"step": 85
|
136 |
},
|
137 |
{
|
138 |
-
"epoch": 0.
|
139 |
-
"grad_norm": 2.
|
140 |
-
"learning_rate": 7.
|
141 |
-
"loss": 0.
|
142 |
"step": 90
|
143 |
},
|
144 |
{
|
145 |
-
"epoch": 0.
|
146 |
-
"grad_norm": 2.
|
147 |
-
"learning_rate": 7.
|
148 |
-
"loss": 0.
|
149 |
"step": 95
|
150 |
},
|
151 |
{
|
152 |
-
"epoch": 0.
|
153 |
-
"grad_norm": 2.
|
154 |
-
"learning_rate":
|
155 |
-
"loss": 0.
|
156 |
"step": 100
|
157 |
},
|
158 |
{
|
159 |
-
"epoch": 0.
|
160 |
-
"grad_norm": 2.
|
161 |
-
"learning_rate": 6.
|
162 |
-
"loss": 0.
|
163 |
"step": 105
|
164 |
},
|
165 |
{
|
166 |
-
"epoch": 0.
|
167 |
-
"grad_norm": 2.
|
168 |
-
"learning_rate": 6.
|
169 |
-
"loss": 0.
|
170 |
"step": 110
|
171 |
},
|
172 |
{
|
173 |
-
"epoch": 0.
|
174 |
-
"grad_norm": 2.
|
175 |
-
"learning_rate":
|
176 |
-
"loss": 0.
|
177 |
"step": 115
|
178 |
},
|
179 |
{
|
180 |
-
"epoch": 0.
|
181 |
-
"grad_norm": 2.
|
182 |
-
"learning_rate": 5.
|
183 |
-
"loss": 0.
|
184 |
"step": 120
|
185 |
},
|
186 |
{
|
187 |
-
"epoch": 0.
|
188 |
-
"grad_norm": 2.
|
189 |
-
"learning_rate": 5.
|
190 |
-
"loss": 0.
|
191 |
"step": 125
|
192 |
},
|
193 |
{
|
194 |
-
"epoch": 0.
|
195 |
-
"grad_norm": 2.
|
196 |
-
"learning_rate": 4.
|
197 |
-
"loss": 0.
|
198 |
"step": 130
|
199 |
},
|
200 |
{
|
201 |
-
"epoch": 0.
|
202 |
-
"grad_norm": 2.
|
203 |
-
"learning_rate": 4.
|
204 |
-
"loss": 0.
|
205 |
"step": 135
|
206 |
},
|
207 |
{
|
208 |
-
"epoch": 0.
|
209 |
-
"grad_norm": 2.
|
210 |
-
"learning_rate":
|
211 |
-
"loss": 0.
|
212 |
"step": 140
|
213 |
},
|
214 |
{
|
215 |
-
"epoch": 0.
|
216 |
-
"grad_norm": 2.
|
217 |
-
"learning_rate": 3.
|
218 |
-
"loss": 0.
|
219 |
"step": 145
|
220 |
},
|
221 |
{
|
222 |
-
"epoch": 0.
|
223 |
-
"grad_norm": 2.
|
224 |
-
"learning_rate": 3.
|
225 |
-
"loss": 0.
|
226 |
"step": 150
|
227 |
},
|
228 |
{
|
229 |
-
"epoch": 0.
|
230 |
-
"grad_norm": 2.
|
231 |
-
"learning_rate":
|
232 |
-
"loss": 0.
|
233 |
"step": 155
|
234 |
},
|
235 |
{
|
236 |
-
"epoch": 0.
|
237 |
-
"grad_norm": 2.
|
238 |
-
"learning_rate": 2.
|
239 |
-
"loss": 0.
|
240 |
"step": 160
|
241 |
},
|
242 |
{
|
243 |
-
"epoch": 0.
|
244 |
-
"grad_norm": 2.
|
245 |
-
"learning_rate": 2.
|
246 |
-
"loss": 0.
|
247 |
"step": 165
|
248 |
},
|
249 |
{
|
250 |
-
"epoch": 0.
|
251 |
-
"grad_norm": 2.
|
252 |
-
"learning_rate":
|
253 |
-
"loss": 0.
|
254 |
"step": 170
|
255 |
},
|
256 |
{
|
257 |
-
"epoch": 0.
|
258 |
-
"grad_norm": 2.
|
259 |
-
"learning_rate": 1.
|
260 |
-
"loss": 0.
|
261 |
"step": 175
|
262 |
},
|
263 |
{
|
264 |
-
"epoch": 0.
|
265 |
-
"grad_norm": 2.
|
266 |
-
"learning_rate": 1.
|
267 |
-
"loss": 0.
|
268 |
"step": 180
|
269 |
},
|
270 |
{
|
271 |
-
"epoch": 0.
|
272 |
-
"grad_norm": 2.
|
273 |
-
"learning_rate": 1.
|
274 |
-
"loss": 0.
|
275 |
"step": 185
|
276 |
},
|
277 |
{
|
278 |
-
"epoch": 0.
|
279 |
-
"grad_norm": 2.
|
280 |
-
"learning_rate":
|
281 |
-
"loss": 0.
|
282 |
"step": 190
|
283 |
},
|
284 |
{
|
285 |
-
"epoch": 0.
|
286 |
-
"grad_norm": 2.
|
287 |
-
"learning_rate":
|
288 |
-
"loss": 0.
|
289 |
"step": 195
|
290 |
},
|
291 |
{
|
292 |
-
"epoch": 0.
|
293 |
-
"grad_norm": 2.
|
294 |
-
"learning_rate":
|
295 |
-
"loss": 0.
|
296 |
"step": 200
|
297 |
},
|
298 |
{
|
299 |
-
"epoch": 0.
|
300 |
-
"grad_norm": 2.
|
301 |
-
"learning_rate":
|
302 |
-
"loss": 0.
|
303 |
"step": 205
|
304 |
},
|
305 |
{
|
306 |
-
"epoch": 0.
|
307 |
-
"grad_norm": 2.
|
308 |
-
"learning_rate":
|
309 |
-
"loss": 0.
|
310 |
"step": 210
|
311 |
},
|
312 |
{
|
313 |
-
"epoch": 0.
|
314 |
-
"grad_norm": 2.
|
315 |
-
"learning_rate":
|
316 |
-
"loss": 0.
|
317 |
"step": 215
|
318 |
},
|
319 |
{
|
320 |
-
"epoch": 0.
|
321 |
-
"grad_norm": 2.
|
322 |
-
"learning_rate":
|
323 |
-
"loss": 0.
|
324 |
"step": 220
|
325 |
},
|
326 |
{
|
327 |
-
"epoch": 0.
|
328 |
-
"grad_norm": 2.
|
329 |
-
"learning_rate":
|
330 |
-
"loss": 0.
|
331 |
"step": 225
|
332 |
},
|
333 |
{
|
334 |
-
"epoch": 0.
|
335 |
-
"
|
336 |
-
"
|
337 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
338 |
"eval_steps_per_second": 0.922,
|
339 |
-
"step":
|
340 |
},
|
341 |
{
|
342 |
-
"epoch": 0.
|
343 |
-
"step":
|
344 |
-
"total_flos":
|
345 |
-
"train_loss": 0.
|
346 |
-
"train_runtime":
|
347 |
-
"train_samples_per_second": 1.
|
348 |
"train_steps_per_second": 0.032
|
349 |
}
|
350 |
],
|
351 |
"logging_steps": 5,
|
352 |
-
"max_steps":
|
353 |
"num_input_tokens_seen": 0,
|
354 |
"num_train_epochs": 1,
|
355 |
"save_steps": 100,
|
@@ -365,7 +372,7 @@
|
|
365 |
"attributes": {}
|
366 |
}
|
367 |
},
|
368 |
-
"total_flos":
|
369 |
"train_batch_size": 2,
|
370 |
"trial_name": null,
|
371 |
"trial_params": null
|
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
+
"epoch": 0.997867803837953,
|
5 |
"eval_steps": 500,
|
6 |
+
"global_step": 234,
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
10 |
"log_history": [
|
11 |
{
|
12 |
+
"epoch": 0.0042643923240938165,
|
13 |
+
"grad_norm": 25.654364462455888,
|
14 |
+
"learning_rate": 4.1666666666666667e-07,
|
15 |
+
"loss": 1.5118,
|
16 |
"step": 1
|
17 |
},
|
18 |
{
|
19 |
+
"epoch": 0.021321961620469083,
|
20 |
+
"grad_norm": 9.217679826848354,
|
21 |
+
"learning_rate": 2.0833333333333334e-06,
|
22 |
+
"loss": 1.4588,
|
23 |
"step": 5
|
24 |
},
|
25 |
{
|
26 |
+
"epoch": 0.042643923240938165,
|
27 |
+
"grad_norm": 3.8364568135386343,
|
28 |
+
"learning_rate": 4.166666666666667e-06,
|
29 |
+
"loss": 1.1844,
|
30 |
"step": 10
|
31 |
},
|
32 |
{
|
33 |
+
"epoch": 0.06396588486140725,
|
34 |
+
"grad_norm": 2.7685426084723606,
|
35 |
+
"learning_rate": 6.25e-06,
|
36 |
+
"loss": 1.0266,
|
37 |
"step": 15
|
38 |
},
|
39 |
{
|
40 |
+
"epoch": 0.08528784648187633,
|
41 |
+
"grad_norm": 2.714167254334698,
|
42 |
+
"learning_rate": 8.333333333333334e-06,
|
43 |
+
"loss": 0.9764,
|
44 |
"step": 20
|
45 |
},
|
46 |
{
|
47 |
+
"epoch": 0.10660980810234541,
|
48 |
+
"grad_norm": 2.575043424767955,
|
49 |
+
"learning_rate": 9.999440509051367e-06,
|
50 |
+
"loss": 0.9474,
|
51 |
"step": 25
|
52 |
},
|
53 |
{
|
54 |
+
"epoch": 0.1279317697228145,
|
55 |
+
"grad_norm": 2.5473126644658635,
|
56 |
+
"learning_rate": 9.979871469976197e-06,
|
57 |
+
"loss": 0.9265,
|
58 |
"step": 30
|
59 |
},
|
60 |
{
|
61 |
+
"epoch": 0.14925373134328357,
|
62 |
+
"grad_norm": 2.7728522540985927,
|
63 |
+
"learning_rate": 9.932452969617607e-06,
|
64 |
+
"loss": 0.9103,
|
65 |
"step": 35
|
66 |
},
|
67 |
{
|
68 |
+
"epoch": 0.17057569296375266,
|
69 |
+
"grad_norm": 2.262656802643975,
|
70 |
+
"learning_rate": 9.857450191464337e-06,
|
71 |
+
"loss": 0.9089,
|
72 |
"step": 40
|
73 |
},
|
74 |
{
|
75 |
+
"epoch": 0.19189765458422176,
|
76 |
+
"grad_norm": 2.2929860595064353,
|
77 |
+
"learning_rate": 9.755282581475769e-06,
|
78 |
+
"loss": 0.8839,
|
79 |
"step": 45
|
80 |
},
|
81 |
{
|
82 |
+
"epoch": 0.21321961620469082,
|
83 |
+
"grad_norm": 2.9962187125117556,
|
84 |
+
"learning_rate": 9.626521502369984e-06,
|
85 |
+
"loss": 0.8779,
|
86 |
"step": 50
|
87 |
},
|
88 |
{
|
89 |
+
"epoch": 0.2345415778251599,
|
90 |
+
"grad_norm": 2.4461853196937744,
|
91 |
+
"learning_rate": 9.471887038331686e-06,
|
92 |
+
"loss": 0.8655,
|
93 |
"step": 55
|
94 |
},
|
95 |
{
|
96 |
+
"epoch": 0.255863539445629,
|
97 |
+
"grad_norm": 2.548713200713329,
|
98 |
+
"learning_rate": 9.292243968009332e-06,
|
99 |
+
"loss": 0.8452,
|
100 |
"step": 60
|
101 |
},
|
102 |
{
|
103 |
+
"epoch": 0.2771855010660981,
|
104 |
+
"grad_norm": 2.354080355646257,
|
105 |
+
"learning_rate": 9.088596928322158e-06,
|
106 |
+
"loss": 0.8453,
|
107 |
"step": 65
|
108 |
},
|
109 |
{
|
110 |
+
"epoch": 0.29850746268656714,
|
111 |
+
"grad_norm": 2.3350186621937494,
|
112 |
+
"learning_rate": 8.862084796122998e-06,
|
113 |
+
"loss": 0.8213,
|
114 |
"step": 70
|
115 |
},
|
116 |
{
|
117 |
+
"epoch": 0.31982942430703626,
|
118 |
+
"grad_norm": 2.352888208422696,
|
119 |
+
"learning_rate": 8.613974319136959e-06,
|
120 |
+
"loss": 0.8087,
|
121 |
"step": 75
|
122 |
},
|
123 |
{
|
124 |
+
"epoch": 0.3411513859275053,
|
125 |
+
"grad_norm": 2.626490865987853,
|
126 |
+
"learning_rate": 8.345653031794292e-06,
|
127 |
+
"loss": 0.8,
|
128 |
"step": 80
|
129 |
},
|
130 |
{
|
131 |
+
"epoch": 0.3624733475479744,
|
132 |
+
"grad_norm": 2.2564126156464934,
|
133 |
+
"learning_rate": 8.058621495575032e-06,
|
134 |
+
"loss": 0.7883,
|
135 |
"step": 85
|
136 |
},
|
137 |
{
|
138 |
+
"epoch": 0.3837953091684435,
|
139 |
+
"grad_norm": 2.536678489630529,
|
140 |
+
"learning_rate": 7.754484907260513e-06,
|
141 |
+
"loss": 0.7797,
|
142 |
"step": 90
|
143 |
},
|
144 |
{
|
145 |
+
"epoch": 0.4051172707889126,
|
146 |
+
"grad_norm": 2.330261835490306,
|
147 |
+
"learning_rate": 7.434944122021837e-06,
|
148 |
+
"loss": 0.7704,
|
149 |
"step": 95
|
150 |
},
|
151 |
{
|
152 |
+
"epoch": 0.42643923240938164,
|
153 |
+
"grad_norm": 2.375473887900136,
|
154 |
+
"learning_rate": 7.101786141547829e-06,
|
155 |
+
"loss": 0.7491,
|
156 |
"step": 100
|
157 |
},
|
158 |
{
|
159 |
+
"epoch": 0.44776119402985076,
|
160 |
+
"grad_norm": 2.31845485895562,
|
161 |
+
"learning_rate": 6.7568741204067145e-06,
|
162 |
+
"loss": 0.7426,
|
163 |
"step": 105
|
164 |
},
|
165 |
{
|
166 |
+
"epoch": 0.4690831556503198,
|
167 |
+
"grad_norm": 2.2326175780721513,
|
168 |
+
"learning_rate": 6.402136946530014e-06,
|
169 |
+
"loss": 0.7366,
|
170 |
"step": 110
|
171 |
},
|
172 |
{
|
173 |
+
"epoch": 0.4904051172707889,
|
174 |
+
"grad_norm": 2.444799836226394,
|
175 |
+
"learning_rate": 6.039558454088796e-06,
|
176 |
+
"loss": 0.7294,
|
177 |
"step": 115
|
178 |
},
|
179 |
{
|
180 |
+
"epoch": 0.511727078891258,
|
181 |
+
"grad_norm": 2.42023799653421,
|
182 |
+
"learning_rate": 5.671166329088278e-06,
|
183 |
+
"loss": 0.7346,
|
184 |
"step": 120
|
185 |
},
|
186 |
{
|
187 |
+
"epoch": 0.5330490405117271,
|
188 |
+
"grad_norm": 2.525769921790198,
|
189 |
+
"learning_rate": 5.299020769725172e-06,
|
190 |
+
"loss": 0.716,
|
191 |
"step": 125
|
192 |
},
|
193 |
{
|
194 |
+
"epoch": 0.5543710021321961,
|
195 |
+
"grad_norm": 2.210624855154462,
|
196 |
+
"learning_rate": 4.9252029649236835e-06,
|
197 |
+
"loss": 0.7087,
|
198 |
"step": 130
|
199 |
},
|
200 |
{
|
201 |
+
"epoch": 0.5756929637526652,
|
202 |
+
"grad_norm": 2.260417777455262,
|
203 |
+
"learning_rate": 4.551803455482833e-06,
|
204 |
+
"loss": 0.6979,
|
205 |
"step": 135
|
206 |
},
|
207 |
{
|
208 |
+
"epoch": 0.5970149253731343,
|
209 |
+
"grad_norm": 2.5410734519213847,
|
210 |
+
"learning_rate": 4.180910442924312e-06,
|
211 |
+
"loss": 0.6758,
|
212 |
"step": 140
|
213 |
},
|
214 |
{
|
215 |
+
"epoch": 0.6183368869936035,
|
216 |
+
"grad_norm": 2.2197214614990983,
|
217 |
+
"learning_rate": 3.8145981114225135e-06,
|
218 |
+
"loss": 0.6832,
|
219 |
"step": 145
|
220 |
},
|
221 |
{
|
222 |
+
"epoch": 0.6396588486140725,
|
223 |
+
"grad_norm": 2.417478197312417,
|
224 |
+
"learning_rate": 3.4549150281252635e-06,
|
225 |
+
"loss": 0.6705,
|
226 |
"step": 150
|
227 |
},
|
228 |
{
|
229 |
+
"epoch": 0.6609808102345416,
|
230 |
+
"grad_norm": 2.193206874567919,
|
231 |
+
"learning_rate": 3.1038726867353587e-06,
|
232 |
+
"loss": 0.6909,
|
233 |
"step": 155
|
234 |
},
|
235 |
{
|
236 |
+
"epoch": 0.6823027718550106,
|
237 |
+
"grad_norm": 2.3141978562259133,
|
238 |
+
"learning_rate": 2.7634342584218364e-06,
|
239 |
+
"loss": 0.678,
|
240 |
"step": 160
|
241 |
},
|
242 |
{
|
243 |
+
"epoch": 0.7036247334754797,
|
244 |
+
"grad_norm": 2.20282691421215,
|
245 |
+
"learning_rate": 2.43550361297047e-06,
|
246 |
+
"loss": 0.6646,
|
247 |
"step": 165
|
248 |
},
|
249 |
{
|
250 |
+
"epoch": 0.7249466950959488,
|
251 |
+
"grad_norm": 2.3241432733966962,
|
252 |
+
"learning_rate": 2.1219146715716332e-06,
|
253 |
+
"loss": 0.6633,
|
254 |
"step": 170
|
255 |
},
|
256 |
{
|
257 |
+
"epoch": 0.746268656716418,
|
258 |
+
"grad_norm": 2.3658483418520464,
|
259 |
+
"learning_rate": 1.8244211507891064e-06,
|
260 |
+
"loss": 0.6516,
|
261 |
"step": 175
|
262 |
},
|
263 |
{
|
264 |
+
"epoch": 0.767590618336887,
|
265 |
+
"grad_norm": 2.259696417637488,
|
266 |
+
"learning_rate": 1.544686755065677e-06,
|
267 |
+
"loss": 0.6418,
|
268 |
"step": 180
|
269 |
},
|
270 |
{
|
271 |
+
"epoch": 0.7889125799573561,
|
272 |
+
"grad_norm": 2.284368925546414,
|
273 |
+
"learning_rate": 1.2842758726130283e-06,
|
274 |
+
"loss": 0.6405,
|
275 |
"step": 185
|
276 |
},
|
277 |
{
|
278 |
+
"epoch": 0.8102345415778252,
|
279 |
+
"grad_norm": 2.2174015564488223,
|
280 |
+
"learning_rate": 1.044644826718295e-06,
|
281 |
+
"loss": 0.6359,
|
282 |
"step": 190
|
283 |
},
|
284 |
{
|
285 |
+
"epoch": 0.8315565031982942,
|
286 |
+
"grad_norm": 2.3098966859462076,
|
287 |
+
"learning_rate": 8.271337313934869e-07,
|
288 |
+
"loss": 0.6232,
|
289 |
"step": 195
|
290 |
},
|
291 |
{
|
292 |
+
"epoch": 0.8528784648187633,
|
293 |
+
"grad_norm": 2.240425165408693,
|
294 |
+
"learning_rate": 6.329589969143518e-07,
|
295 |
+
"loss": 0.6263,
|
296 |
"step": 200
|
297 |
},
|
298 |
{
|
299 |
+
"epoch": 0.8742004264392325,
|
300 |
+
"grad_norm": 2.203409177091297,
|
301 |
+
"learning_rate": 4.632065271606756e-07,
|
302 |
+
"loss": 0.6299,
|
303 |
"step": 205
|
304 |
},
|
305 |
{
|
306 |
+
"epoch": 0.8955223880597015,
|
307 |
+
"grad_norm": 2.1702011902470724,
|
308 |
+
"learning_rate": 3.18825646801314e-07,
|
309 |
+
"loss": 0.636,
|
310 |
"step": 210
|
311 |
},
|
312 |
{
|
313 |
+
"epoch": 0.9168443496801706,
|
314 |
+
"grad_norm": 2.202446820245564,
|
315 |
+
"learning_rate": 2.006237922855553e-07,
|
316 |
+
"loss": 0.6182,
|
317 |
"step": 215
|
318 |
},
|
319 |
{
|
320 |
+
"epoch": 0.9381663113006397,
|
321 |
+
"grad_norm": 2.118840248626809,
|
322 |
+
"learning_rate": 1.0926199633097156e-07,
|
323 |
+
"loss": 0.609,
|
324 |
"step": 220
|
325 |
},
|
326 |
{
|
327 |
+
"epoch": 0.9594882729211087,
|
328 |
+
"grad_norm": 2.168175873632397,
|
329 |
+
"learning_rate": 4.52511911603265e-08,
|
330 |
+
"loss": 0.6173,
|
331 |
"step": 225
|
332 |
},
|
333 |
{
|
334 |
+
"epoch": 0.9808102345415778,
|
335 |
+
"grad_norm": 2.2624619803066617,
|
336 |
+
"learning_rate": 8.949351161324227e-09,
|
337 |
+
"loss": 0.6207,
|
338 |
+
"step": 230
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.997867803837953,
|
342 |
+
"eval_loss": 0.7390011548995972,
|
343 |
+
"eval_runtime": 106.273,
|
344 |
+
"eval_samples_per_second": 3.67,
|
345 |
"eval_steps_per_second": 0.922,
|
346 |
+
"step": 234
|
347 |
},
|
348 |
{
|
349 |
+
"epoch": 0.997867803837953,
|
350 |
+
"step": 234,
|
351 |
+
"total_flos": 48942494515200.0,
|
352 |
+
"train_loss": 0.7695284368645432,
|
353 |
+
"train_runtime": 7306.9109,
|
354 |
+
"train_samples_per_second": 1.026,
|
355 |
"train_steps_per_second": 0.032
|
356 |
}
|
357 |
],
|
358 |
"logging_steps": 5,
|
359 |
+
"max_steps": 234,
|
360 |
"num_input_tokens_seen": 0,
|
361 |
"num_train_epochs": 1,
|
362 |
"save_steps": 100,
|
|
|
372 |
"attributes": {}
|
373 |
}
|
374 |
},
|
375 |
+
"total_flos": 48942494515200.0,
|
376 |
"train_batch_size": 2,
|
377 |
"trial_name": null,
|
378 |
"trial_params": null
|
training_args.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 6520
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:53ce655e7ab1129bcd883d7974bca1f5c5778062e5c8a4eb41ef38796a8ee647
|
3 |
size 6520
|