Hugging Face's logo

EllieS
/
zephyr-dpo-timedial-selfgen-mix2

PEFT
TensorBoard
Safetensors
mistral
alignment-handbook
trl
dpo
Generated from Trainer
Model card Files Metrics Community
EllieS commited on
Commit
669ef4d
•
1 Parent(s): 776f195

Model save

Browse files
Files changed (6) hide show
  1. README.md +74 -0
  2. adapter_model.safetensors +1 -1
  3. all_results.json +8 -0
  4. runs/Apr12_04-33-13_612e66badb5c/events.out.tfevents.1712896524.612e66badb5c.12978.0 +2 -2
  5. train_results.json +8 -0
  6. trainer_state.json +497 -0
README.md ADDED
@@ -0,0 +1,74 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ library_name: peft
4
+ tags:
5
+ - trl
6
+ - dpo
7
+ - generated_from_trainer
8
+ base_model: alignment-handbook/zephyr-7b-sft-full
9
+ model-index:
10
+ - name: zephyr-dpo-timedial-selfgen-mix2
11
+ results: []
12
+ ---
13
+
14
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
15
+ should probably proofread and complete it, then remove this comment. -->
16
+
17
+ # zephyr-dpo-timedial-selfgen-mix2
18
+
19
+ This model is a fine-tuned version of [alignment-handbook/zephyr-7b-sft-full](https://huggingface.co/alignment-handbook/zephyr-7b-sft-full) on an unknown dataset.
20
+ It achieves the following results on the evaluation set:
21
+ - Loss: 0.2588
22
+ - Rewards/chosen: 0.2087
23
+ - Rewards/rejected: -1.0200
24
+ - Rewards/accuracies: 1.0
25
+ - Rewards/margins: 1.2287
26
+ - Logps/rejected: -132.2824
27
+ - Logps/chosen: -6.8403
28
+ - Logits/rejected: -2.8679
29
+ - Logits/chosen: -2.8807
30
+
31
+ ## Model description
32
+
33
+ More information needed
34
+
35
+ ## Intended uses & limitations
36
+
37
+ More information needed
38
+
39
+ ## Training and evaluation data
40
+
41
+ More information needed
42
+
43
+ ## Training procedure
44
+
45
+ ### Training hyperparameters
46
+
47
+ The following hyperparameters were used during training:
48
+ - learning_rate: 5e-06
49
+ - train_batch_size: 2
50
+ - eval_batch_size: 2
51
+ - seed: 42
52
+ - distributed_type: multi-GPU
53
+ - gradient_accumulation_steps: 2
54
+ - total_train_batch_size: 4
55
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
56
+ - lr_scheduler_type: cosine
57
+ - lr_scheduler_warmup_ratio: 0.1
58
+ - num_epochs: 1
59
+
60
+ ### Training results
61
+
62
+ | Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
63
+ |:-------------:|:-----:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|
64
+ | 0.4825 | 0.35 | 100 | 0.4632 | 0.2303 | -0.3022 | 1.0 | 0.5325 | -60.5070 | -4.6766 | -2.9751 | -2.9715 |
65
+ | 0.2786 | 0.69 | 200 | 0.2588 | 0.2087 | -1.0200 | 1.0 | 1.2287 | -132.2824 | -6.8403 | -2.8679 | -2.8807 |
66
+
67
+
68
+ ### Framework versions
69
+
70
+ - PEFT 0.7.1
71
+ - Transformers 4.39.0.dev0
72
+ - Pytorch 2.1.2+cu121
73
+ - Datasets 2.18.0
74
+ - Tokenizers 0.15.2
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:59dbdcba02b934a32b46914a0d25927a5c67bb12323bb67cac04d4eb94b605b4
3
  size 83946192
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9964ad62a8bf5a21fb9b027f8ce26c447d3fa36afcd73ed4cfcdd37f63636b91
3
  size 83946192
all_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "train_loss": 0.4018066387275511,
4
+ "train_runtime": 624.4981,
5
+ "train_samples": 1157,
6
+ "train_samples_per_second": 1.853,
7
+ "train_steps_per_second": 0.463
8
+ }
runs/Apr12_04-33-13_612e66badb5c/events.out.tfevents.1712896524.612e66badb5c.12978.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:20c094e0e641ed70f5394c79937502bd4cd1cc6d3cc1c43341ae8d344e231a1c
3
- size 20577
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6e9e6913d29b2073ae56e076d61ca361218f3d3c405a65986b72c7d1c6984931
3
+ size 26435
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "train_loss": 0.4018066387275511,
4
+ "train_runtime": 624.4981,
5
+ "train_samples": 1157,
6
+ "train_samples_per_second": 1.853,
7
+ "train_steps_per_second": 0.463
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,497 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.998272884283247,
5
+ "eval_steps": 100,
6
+ "global_step": 289,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0,
13
+ "grad_norm": 0.2451171875,
14
+ "learning_rate": 1.7241379310344828e-07,
15
+ "logits/chosen": -2.946108341217041,
16
+ "logits/rejected": -2.9481201171875,
17
+ "logps/chosen": -27.542381286621094,
18
+ "logps/rejected": -28.998493194580078,
19
+ "loss": 0.6931,
20
+ "rewards/accuracies": 0.0,
21
+ "rewards/chosen": 0.0,
22
+ "rewards/margins": 0.0,
23
+ "rewards/rejected": 0.0,
24
+ "step": 1
25
+ },
26
+ {
27
+ "epoch": 0.03,
28
+ "grad_norm": 0.2431640625,
29
+ "learning_rate": 1.724137931034483e-06,
30
+ "logits/chosen": -2.967113494873047,
31
+ "logits/rejected": -2.9693994522094727,
32
+ "logps/chosen": -27.679420471191406,
33
+ "logps/rejected": -30.345211029052734,
34
+ "loss": 0.6931,
35
+ "rewards/accuracies": 0.4166666567325592,
36
+ "rewards/chosen": -0.00010507476690690964,
37
+ "rewards/margins": 6.562391354236752e-05,
38
+ "rewards/rejected": -0.0001706986950011924,
39
+ "step": 10
40
+ },
41
+ {
42
+ "epoch": 0.07,
43
+ "grad_norm": 0.259765625,
44
+ "learning_rate": 3.448275862068966e-06,
45
+ "logits/chosen": -2.9687092304229736,
46
+ "logits/rejected": -2.9725170135498047,
47
+ "logps/chosen": -27.916067123413086,
48
+ "logps/rejected": -32.21611404418945,
49
+ "loss": 0.6923,
50
+ "rewards/accuracies": 0.800000011920929,
51
+ "rewards/chosen": 0.0016102585941553116,
52
+ "rewards/margins": 0.0016238940879702568,
53
+ "rewards/rejected": -1.3635644791065715e-05,
54
+ "step": 20
55
+ },
56
+ {
57
+ "epoch": 0.1,
58
+ "grad_norm": 0.263671875,
59
+ "learning_rate": 4.999817502139027e-06,
60
+ "logits/chosen": -2.991443157196045,
61
+ "logits/rejected": -2.992274761199951,
62
+ "logps/chosen": -27.992717742919922,
63
+ "logps/rejected": -30.998449325561523,
64
+ "loss": 0.6894,
65
+ "rewards/accuracies": 0.9750000238418579,
66
+ "rewards/chosen": 0.010343936271965504,
67
+ "rewards/margins": 0.007500241510570049,
68
+ "rewards/rejected": 0.002843693597242236,
69
+ "step": 30
70
+ },
71
+ {
72
+ "epoch": 0.14,
73
+ "grad_norm": 0.359375,
74
+ "learning_rate": 4.977949980164773e-06,
75
+ "logits/chosen": -2.9539809226989746,
76
+ "logits/rejected": -2.9519870281219482,
77
+ "logps/chosen": -26.01015853881836,
78
+ "logps/rejected": -30.401302337646484,
79
+ "loss": 0.6829,
80
+ "rewards/accuracies": 1.0,
81
+ "rewards/chosen": 0.03149240463972092,
82
+ "rewards/margins": 0.02063887007534504,
83
+ "rewards/rejected": 0.010853536427021027,
84
+ "step": 40
85
+ },
86
+ {
87
+ "epoch": 0.17,
88
+ "grad_norm": 0.34375,
89
+ "learning_rate": 4.919948367622307e-06,
90
+ "logits/chosen": -2.983638048171997,
91
+ "logits/rejected": -2.989013195037842,
92
+ "logps/chosen": -22.443767547607422,
93
+ "logps/rejected": -31.806631088256836,
94
+ "loss": 0.6664,
95
+ "rewards/accuracies": 1.0,
96
+ "rewards/chosen": 0.06118880584836006,
97
+ "rewards/margins": 0.05425649881362915,
98
+ "rewards/rejected": 0.006932304706424475,
99
+ "step": 50
100
+ },
101
+ {
102
+ "epoch": 0.21,
103
+ "grad_norm": 0.484375,
104
+ "learning_rate": 4.8266584586307555e-06,
105
+ "logits/chosen": -2.988079786300659,
106
+ "logits/rejected": -2.9914956092834473,
107
+ "logps/chosen": -19.77129364013672,
108
+ "logps/rejected": -29.180835723876953,
109
+ "loss": 0.6535,
110
+ "rewards/accuracies": 0.9750000238418579,
111
+ "rewards/chosen": 0.08603720366954803,
112
+ "rewards/margins": 0.08121357858181,
113
+ "rewards/rejected": 0.004823616705834866,
114
+ "step": 60
115
+ },
116
+ {
117
+ "epoch": 0.24,
118
+ "grad_norm": 0.8515625,
119
+ "learning_rate": 4.699440630133794e-06,
120
+ "logits/chosen": -2.995100736618042,
121
+ "logits/rejected": -2.99615216255188,
122
+ "logps/chosen": -17.9371395111084,
123
+ "logps/rejected": -34.77935028076172,
124
+ "loss": 0.6236,
125
+ "rewards/accuracies": 1.0,
126
+ "rewards/chosen": 0.11037828773260117,
127
+ "rewards/margins": 0.14534664154052734,
128
+ "rewards/rejected": -0.03496835380792618,
129
+ "step": 70
130
+ },
131
+ {
132
+ "epoch": 0.28,
133
+ "grad_norm": 0.84765625,
134
+ "learning_rate": 4.5401500045405126e-06,
135
+ "logits/chosen": -2.9611220359802246,
136
+ "logits/rejected": -2.963106393814087,
137
+ "logps/chosen": -11.03026008605957,
138
+ "logps/rejected": -39.37724304199219,
139
+ "loss": 0.5669,
140
+ "rewards/accuracies": 1.0,
141
+ "rewards/chosen": 0.18076485395431519,
142
+ "rewards/margins": 0.27246183156967163,
143
+ "rewards/rejected": -0.09169697761535645,
144
+ "step": 80
145
+ },
146
+ {
147
+ "epoch": 0.31,
148
+ "grad_norm": 0.59765625,
149
+ "learning_rate": 4.351109397863309e-06,
150
+ "logits/chosen": -2.970407009124756,
151
+ "logits/rejected": -2.9756150245666504,
152
+ "logps/chosen": -6.9094672203063965,
153
+ "logps/rejected": -49.88671112060547,
154
+ "loss": 0.5168,
155
+ "rewards/accuracies": 1.0,
156
+ "rewards/chosen": 0.20805075764656067,
157
+ "rewards/margins": 0.3929779827594757,
158
+ "rewards/rejected": -0.18492721021175385,
159
+ "step": 90
160
+ },
161
+ {
162
+ "epoch": 0.35,
163
+ "grad_norm": 0.828125,
164
+ "learning_rate": 4.135075447829912e-06,
165
+ "logits/chosen": -2.966463327407837,
166
+ "logits/rejected": -2.973740816116333,
167
+ "logps/chosen": -5.979574203491211,
168
+ "logps/rejected": -56.75745391845703,
169
+ "loss": 0.4825,
170
+ "rewards/accuracies": 1.0,
171
+ "rewards/chosen": 0.23063524067401886,
172
+ "rewards/margins": 0.4814692437648773,
173
+ "rewards/rejected": -0.25083404779434204,
174
+ "step": 100
175
+ },
176
+ {
177
+ "epoch": 0.35,
178
+ "eval_logits/chosen": -2.971499443054199,
179
+ "eval_logits/rejected": -2.975092887878418,
180
+ "eval_logps/chosen": -4.676602840423584,
181
+ "eval_logps/rejected": -60.5069694519043,
182
+ "eval_loss": 0.46322527527809143,
183
+ "eval_rewards/accuracies": 1.0,
184
+ "eval_rewards/chosen": 0.2303335815668106,
185
+ "eval_rewards/margins": 0.5325379371643066,
186
+ "eval_rewards/rejected": -0.3022043704986572,
187
+ "eval_runtime": 59.0284,
188
+ "eval_samples_per_second": 4.896,
189
+ "eval_steps_per_second": 2.456,
190
+ "step": 100
191
+ },
192
+ {
193
+ "epoch": 0.38,
194
+ "grad_norm": 2.15625,
195
+ "learning_rate": 3.895198415897896e-06,
196
+ "logits/chosen": -2.944152355194092,
197
+ "logits/rejected": -2.948521614074707,
198
+ "logps/chosen": -4.758464813232422,
199
+ "logps/rejected": -65.92622375488281,
200
+ "loss": 0.4433,
201
+ "rewards/accuracies": 1.0,
202
+ "rewards/chosen": 0.23539750277996063,
203
+ "rewards/margins": 0.5871593356132507,
204
+ "rewards/rejected": -0.3517618179321289,
205
+ "step": 110
206
+ },
207
+ {
208
+ "epoch": 0.41,
209
+ "grad_norm": 0.515625,
210
+ "learning_rate": 3.634976249348867e-06,
211
+ "logits/chosen": -2.956526279449463,
212
+ "logits/rejected": -2.9596874713897705,
213
+ "logps/chosen": -3.4839115142822266,
214
+ "logps/rejected": -74.23634338378906,
215
+ "loss": 0.4105,
216
+ "rewards/accuracies": 1.0,
217
+ "rewards/chosen": 0.2543177604675293,
218
+ "rewards/margins": 0.6815550923347473,
219
+ "rewards/rejected": -0.427237331867218,
220
+ "step": 120
221
+ },
222
+ {
223
+ "epoch": 0.45,
224
+ "grad_norm": 0.65625,
225
+ "learning_rate": 3.3582035733403963e-06,
226
+ "logits/chosen": -2.9395928382873535,
227
+ "logits/rejected": -2.9400253295898438,
228
+ "logps/chosen": -3.3433098793029785,
229
+ "logps/rejected": -81.93199920654297,
230
+ "loss": 0.384,
231
+ "rewards/accuracies": 1.0,
232
+ "rewards/chosen": 0.25131431221961975,
233
+ "rewards/margins": 0.7623029947280884,
234
+ "rewards/rejected": -0.5109885931015015,
235
+ "step": 130
236
+ },
237
+ {
238
+ "epoch": 0.48,
239
+ "grad_norm": 0.56640625,
240
+ "learning_rate": 3.068916356726475e-06,
241
+ "logits/chosen": -2.9392318725585938,
242
+ "logits/rejected": -2.939736843109131,
243
+ "logps/chosen": -2.7690482139587402,
244
+ "logps/rejected": -91.77839660644531,
245
+ "loss": 0.3548,
246
+ "rewards/accuracies": 1.0,
247
+ "rewards/chosen": 0.24845953285694122,
248
+ "rewards/margins": 0.8564491271972656,
249
+ "rewards/rejected": -0.607989490032196,
250
+ "step": 140
251
+ },
252
+ {
253
+ "epoch": 0.52,
254
+ "grad_norm": 0.52734375,
255
+ "learning_rate": 2.771333058543416e-06,
256
+ "logits/chosen": -2.9237265586853027,
257
+ "logits/rejected": -2.9198527336120605,
258
+ "logps/chosen": -3.30281138420105,
259
+ "logps/rejected": -96.4010009765625,
260
+ "loss": 0.3403,
261
+ "rewards/accuracies": 1.0,
262
+ "rewards/chosen": 0.2496950924396515,
263
+ "rewards/margins": 0.9052786827087402,
264
+ "rewards/rejected": -0.6555835604667664,
265
+ "step": 150
266
+ },
267
+ {
268
+ "epoch": 0.55,
269
+ "grad_norm": 0.50390625,
270
+ "learning_rate": 2.4697931133779566e-06,
271
+ "logits/chosen": -2.9088094234466553,
272
+ "logits/rejected": -2.9039511680603027,
273
+ "logps/chosen": -3.719200849533081,
274
+ "logps/rejected": -103.43408203125,
275
+ "loss": 0.3218,
276
+ "rewards/accuracies": 1.0,
277
+ "rewards/chosen": 0.24762046337127686,
278
+ "rewards/margins": 0.9735357165336609,
279
+ "rewards/rejected": -0.7259151935577393,
280
+ "step": 160
281
+ },
282
+ {
283
+ "epoch": 0.59,
284
+ "grad_norm": 0.96875,
285
+ "learning_rate": 2.168693652639432e-06,
286
+ "logits/chosen": -2.9292285442352295,
287
+ "logits/rejected": -2.9254345893859863,
288
+ "logps/chosen": -1.5337612628936768,
289
+ "logps/rejected": -105.90274810791016,
290
+ "loss": 0.3082,
291
+ "rewards/accuracies": 1.0,
292
+ "rewards/chosen": 0.2660350203514099,
293
+ "rewards/margins": 1.021663784980774,
294
+ "rewards/rejected": -0.7556287050247192,
295
+ "step": 170
296
+ },
297
+ {
298
+ "epoch": 0.62,
299
+ "grad_norm": 0.5546875,
300
+ "learning_rate": 1.872425384482346e-06,
301
+ "logits/chosen": -2.9200375080108643,
302
+ "logits/rejected": -2.9096457958221436,
303
+ "logps/chosen": -1.4461075067520142,
304
+ "logps/rejected": -111.00732421875,
305
+ "loss": 0.2964,
306
+ "rewards/accuracies": 1.0,
307
+ "rewards/chosen": 0.2637704610824585,
308
+ "rewards/margins": 1.0691814422607422,
309
+ "rewards/rejected": -0.8054108619689941,
310
+ "step": 180
311
+ },
312
+ {
313
+ "epoch": 0.66,
314
+ "grad_norm": 2.6875,
315
+ "learning_rate": 1.5853085673944695e-06,
316
+ "logits/chosen": -2.9019381999969482,
317
+ "logits/rejected": -2.889967441558838,
318
+ "logps/chosen": -3.121753692626953,
319
+ "logps/rejected": -114.66499328613281,
320
+ "loss": 0.2975,
321
+ "rewards/accuracies": 1.0,
322
+ "rewards/chosen": 0.24539904296398163,
323
+ "rewards/margins": 1.0781965255737305,
324
+ "rewards/rejected": -0.8327974081039429,
325
+ "step": 190
326
+ },
327
+ {
328
+ "epoch": 0.69,
329
+ "grad_norm": 1.0859375,
330
+ "learning_rate": 1.3115300110997097e-06,
331
+ "logits/chosen": -2.867145538330078,
332
+ "logits/rejected": -2.857390880584717,
333
+ "logps/chosen": -4.095256805419922,
334
+ "logps/rejected": -123.1200180053711,
335
+ "loss": 0.2786,
336
+ "rewards/accuracies": 1.0,
337
+ "rewards/chosen": 0.24890461564064026,
338
+ "rewards/margins": 1.1587778329849243,
339
+ "rewards/rejected": -0.9098731875419617,
340
+ "step": 200
341
+ },
342
+ {
343
+ "epoch": 0.69,
344
+ "eval_logits/chosen": -2.8807153701782227,
345
+ "eval_logits/rejected": -2.8679380416870117,
346
+ "eval_logps/chosen": -6.8403167724609375,
347
+ "eval_logps/rejected": -132.28236389160156,
348
+ "eval_loss": 0.2587815523147583,
349
+ "eval_rewards/accuracies": 1.0,
350
+ "eval_rewards/chosen": 0.2086964100599289,
351
+ "eval_rewards/margins": 1.2286547422409058,
352
+ "eval_rewards/rejected": -1.019958257675171,
353
+ "eval_runtime": 59.0007,
354
+ "eval_samples_per_second": 4.898,
355
+ "eval_steps_per_second": 2.458,
356
+ "step": 200
357
+ },
358
+ {
359
+ "epoch": 0.73,
360
+ "grad_norm": 1.0390625,
361
+ "learning_rate": 1.0550820234444627e-06,
362
+ "logits/chosen": -2.8844265937805176,
363
+ "logits/rejected": -2.8720266819000244,
364
+ "logps/chosen": -7.328047752380371,
365
+ "logps/rejected": -136.2236785888672,
366
+ "loss": 0.2536,
367
+ "rewards/accuracies": 1.0,
368
+ "rewards/chosen": 0.2088628113269806,
369
+ "rewards/margins": 1.2624285221099854,
370
+ "rewards/rejected": -1.0535659790039062,
371
+ "step": 210
372
+ },
373
+ {
374
+ "epoch": 0.76,
375
+ "grad_norm": 1.5,
376
+ "learning_rate": 8.197041935593181e-07,
377
+ "logits/chosen": -2.865748643875122,
378
+ "logits/rejected": -2.8470582962036133,
379
+ "logps/chosen": -3.5205326080322266,
380
+ "logps/rejected": -143.3411865234375,
381
+ "loss": 0.2291,
382
+ "rewards/accuracies": 1.0,
383
+ "rewards/chosen": 0.2387482225894928,
384
+ "rewards/margins": 1.365382432937622,
385
+ "rewards/rejected": -1.126634120941162,
386
+ "step": 220
387
+ },
388
+ {
389
+ "epoch": 0.79,
390
+ "grad_norm": 0.97265625,
391
+ "learning_rate": 6.088288602287159e-07,
392
+ "logits/chosen": -2.8658416271209717,
393
+ "logits/rejected": -2.8466126918792725,
394
+ "logps/chosen": -1.298507809638977,
395
+ "logps/rejected": -147.50143432617188,
396
+ "loss": 0.2162,
397
+ "rewards/accuracies": 1.0,
398
+ "rewards/chosen": 0.2748534083366394,
399
+ "rewards/margins": 1.4337130784988403,
400
+ "rewards/rejected": -1.1588596105575562,
401
+ "step": 230
402
+ },
403
+ {
404
+ "epoch": 0.83,
405
+ "grad_norm": 1.34375,
406
+ "learning_rate": 4.255310606625124e-07,
407
+ "logits/chosen": -2.8704545497894287,
408
+ "logits/rejected": -2.8518197536468506,
409
+ "logps/chosen": -2.5422964096069336,
410
+ "logps/rejected": -151.9074249267578,
411
+ "loss": 0.2081,
412
+ "rewards/accuracies": 1.0,
413
+ "rewards/chosen": 0.25674667954444885,
414
+ "rewards/margins": 1.4753159284591675,
415
+ "rewards/rejected": -1.2185693979263306,
416
+ "step": 240
417
+ },
418
+ {
419
+ "epoch": 0.86,
420
+ "grad_norm": 0.9921875,
421
+ "learning_rate": 2.7248368952908055e-07,
422
+ "logits/chosen": -2.861412525177002,
423
+ "logits/rejected": -2.838165044784546,
424
+ "logps/chosen": -2.5676398277282715,
425
+ "logps/rejected": -152.24969482421875,
426
+ "loss": 0.2105,
427
+ "rewards/accuracies": 1.0,
428
+ "rewards/chosen": 0.2601901888847351,
429
+ "rewards/margins": 1.4616148471832275,
430
+ "rewards/rejected": -1.2014245986938477,
431
+ "step": 250
432
+ },
433
+ {
434
+ "epoch": 0.9,
435
+ "grad_norm": 0.88671875,
436
+ "learning_rate": 1.5191852213221198e-07,
437
+ "logits/chosen": -2.8463876247406006,
438
+ "logits/rejected": -2.817927837371826,
439
+ "logps/chosen": -3.504025936126709,
440
+ "logps/rejected": -156.09738159179688,
441
+ "loss": 0.2054,
442
+ "rewards/accuracies": 1.0,
443
+ "rewards/chosen": 0.2497003823518753,
444
+ "rewards/margins": 1.4921144247055054,
445
+ "rewards/rejected": -1.2424139976501465,
446
+ "step": 260
447
+ },
448
+ {
449
+ "epoch": 0.93,
450
+ "grad_norm": 0.921875,
451
+ "learning_rate": 6.559367010166629e-08,
452
+ "logits/chosen": -2.8501267433166504,
453
+ "logits/rejected": -2.8306031227111816,
454
+ "logps/chosen": -2.110583782196045,
455
+ "logps/rejected": -152.65463256835938,
456
+ "loss": 0.2113,
457
+ "rewards/accuracies": 1.0,
458
+ "rewards/chosen": 0.26196378469467163,
459
+ "rewards/margins": 1.4719316959381104,
460
+ "rewards/rejected": -1.209967851638794,
461
+ "step": 270
462
+ },
463
+ {
464
+ "epoch": 0.97,
465
+ "grad_norm": 1.453125,
466
+ "learning_rate": 1.4767944166687032e-08,
467
+ "logits/chosen": -2.8537425994873047,
468
+ "logits/rejected": -2.827571392059326,
469
+ "logps/chosen": -2.493337869644165,
470
+ "logps/rejected": -154.96142578125,
471
+ "loss": 0.2011,
472
+ "rewards/accuracies": 1.0,
473
+ "rewards/chosen": 0.2604244649410248,
474
+ "rewards/margins": 1.5092506408691406,
475
+ "rewards/rejected": -1.2488261461257935,
476
+ "step": 280
477
+ },
478
+ {
479
+ "epoch": 1.0,
480
+ "step": 289,
481
+ "total_flos": 0.0,
482
+ "train_loss": 0.4018066387275511,
483
+ "train_runtime": 624.4981,
484
+ "train_samples_per_second": 1.853,
485
+ "train_steps_per_second": 0.463
486
+ }
487
+ ],
488
+ "logging_steps": 10,
489
+ "max_steps": 289,
490
+ "num_input_tokens_seen": 0,
491
+ "num_train_epochs": 1,
492
+ "save_steps": 200,
493
+ "total_flos": 0.0,
494
+ "train_batch_size": 2,
495
+ "trial_name": null,
496
+ "trial_params": null
497
+ }