jikaixuan commited on
Commit
cf31064
1 Parent(s): d205170

Training in progress, epoch 0

Browse files
adapter_config.json CHANGED
@@ -19,9 +19,9 @@
19
  "rank_pattern": {},
20
  "revision": null,
21
  "target_modules": [
22
- "q_proj",
23
- "k_proj",
24
  "o_proj",
 
 
25
  "v_proj"
26
  ],
27
  "task_type": "CAUSAL_LM",
 
19
  "rank_pattern": {},
20
  "revision": null,
21
  "target_modules": [
 
 
22
  "o_proj",
23
+ "k_proj",
24
+ "q_proj",
25
  "v_proj"
26
  ],
27
  "task_type": "CAUSAL_LM",
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:447886fe6fad21a81ab28995ca80f14ad58365802ff23a558700892ee76e225c
3
  size 109086672
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e0bd3c234ab6123042ad056d3de8b0b9fb646cba1e541c3b7faafb0d4f8c2e10
3
  size 109086672
all_results.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "eval_logits/chosen": -2.4944536685943604,
4
+ "eval_logits/rejected": -2.4963433742523193,
5
+ "eval_logps/chosen": -269.25555419921875,
6
+ "eval_logps/rejected": -253.21238708496094,
7
+ "eval_loss": 0.6399702429771423,
8
+ "eval_rewards/accuracies": 0.6370000243186951,
9
+ "eval_rewards/chosen": 0.030110126361250877,
10
+ "eval_rewards/margins": 0.05743245780467987,
11
+ "eval_rewards/rejected": -0.027322327718138695,
12
+ "eval_runtime": 803.6977,
13
+ "eval_samples": 2000,
14
+ "eval_samples_per_second": 2.488,
15
+ "eval_steps_per_second": 0.311,
16
+ "train_loss": 0.6598132096035942,
17
+ "train_runtime": 45126.4521,
18
+ "train_samples": 61135,
19
+ "train_samples_per_second": 1.355,
20
+ "train_steps_per_second": 0.021
21
+ }
eval_results.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "eval_logits/chosen": -2.4944536685943604,
4
+ "eval_logits/rejected": -2.4963433742523193,
5
+ "eval_logps/chosen": -269.25555419921875,
6
+ "eval_logps/rejected": -253.21238708496094,
7
+ "eval_loss": 0.6399702429771423,
8
+ "eval_rewards/accuracies": 0.6370000243186951,
9
+ "eval_rewards/chosen": 0.030110126361250877,
10
+ "eval_rewards/margins": 0.05743245780467987,
11
+ "eval_rewards/rejected": -0.027322327718138695,
12
+ "eval_runtime": 803.6977,
13
+ "eval_samples": 2000,
14
+ "eval_samples_per_second": 2.488,
15
+ "eval_steps_per_second": 0.311
16
+ }
runs/Jan10_17-48-24_uclaml03.cs.ucla.edu/events.out.tfevents.1704983954.uclaml03.cs.ucla.edu.1956277.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c156e396a0ae2f31537a0115d6f846f7144f72a6223ae6ccc9f1bd57a938e18d
3
+ size 828
runs/Jan12_09-49-31_uclaml03.cs.ucla.edu/events.out.tfevents.1705082087.uclaml03.cs.ucla.edu.3291235.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d96e80bd0f78334abb6fb7ff942241c12a4657e536a885b74ac4bf06f438286e
3
+ size 66178
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "train_loss": 0.6598132096035942,
4
+ "train_runtime": 45126.4521,
5
+ "train_samples": 61135,
6
+ "train_samples_per_second": 1.355,
7
+ "train_steps_per_second": 0.021
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,1388 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.9997382884061764,
5
+ "eval_steps": 100,
6
+ "global_step": 955,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0,
13
+ "learning_rate": 5.208333333333333e-09,
14
+ "logits/chosen": -2.570180892944336,
15
+ "logits/rejected": -2.5666794776916504,
16
+ "logps/chosen": -302.8643798828125,
17
+ "logps/rejected": -232.7855682373047,
18
+ "loss": 0.6931,
19
+ "rewards/accuracies": 0.0,
20
+ "rewards/chosen": 0.0,
21
+ "rewards/margins": 0.0,
22
+ "rewards/rejected": 0.0,
23
+ "step": 1
24
+ },
25
+ {
26
+ "epoch": 0.01,
27
+ "learning_rate": 5.208333333333333e-08,
28
+ "logits/chosen": -2.547344207763672,
29
+ "logits/rejected": -2.532263994216919,
30
+ "logps/chosen": -294.40087890625,
31
+ "logps/rejected": -276.11126708984375,
32
+ "loss": 0.6879,
33
+ "rewards/accuracies": 0.4340277910232544,
34
+ "rewards/chosen": 0.0004416291194502264,
35
+ "rewards/margins": -0.0003637511981651187,
36
+ "rewards/rejected": 0.000805380754172802,
37
+ "step": 10
38
+ },
39
+ {
40
+ "epoch": 0.02,
41
+ "learning_rate": 1.0416666666666667e-07,
42
+ "logits/chosen": -2.3985350131988525,
43
+ "logits/rejected": -2.426626682281494,
44
+ "logps/chosen": -244.53689575195312,
45
+ "logps/rejected": -248.93624877929688,
46
+ "loss": 0.6882,
47
+ "rewards/accuracies": 0.49687498807907104,
48
+ "rewards/chosen": -0.002964673563838005,
49
+ "rewards/margins": 0.00020738500461447984,
50
+ "rewards/rejected": -0.0031720586121082306,
51
+ "step": 20
52
+ },
53
+ {
54
+ "epoch": 0.03,
55
+ "learning_rate": 1.5624999999999999e-07,
56
+ "logits/chosen": -2.5088179111480713,
57
+ "logits/rejected": -2.4950203895568848,
58
+ "logps/chosen": -265.425537109375,
59
+ "logps/rejected": -244.62191772460938,
60
+ "loss": 0.6868,
61
+ "rewards/accuracies": 0.44999998807907104,
62
+ "rewards/chosen": -0.003862012643367052,
63
+ "rewards/margins": -0.0030291248112916946,
64
+ "rewards/rejected": -0.0008328882977366447,
65
+ "step": 30
66
+ },
67
+ {
68
+ "epoch": 0.04,
69
+ "learning_rate": 2.0833333333333333e-07,
70
+ "logits/chosen": -2.472740888595581,
71
+ "logits/rejected": -2.498608112335205,
72
+ "logps/chosen": -264.92767333984375,
73
+ "logps/rejected": -260.81536865234375,
74
+ "loss": 0.6883,
75
+ "rewards/accuracies": 0.4937500059604645,
76
+ "rewards/chosen": 0.002251622499898076,
77
+ "rewards/margins": 0.002210150007158518,
78
+ "rewards/rejected": 4.1472725570201874e-05,
79
+ "step": 40
80
+ },
81
+ {
82
+ "epoch": 0.05,
83
+ "learning_rate": 2.604166666666667e-07,
84
+ "logits/chosen": -2.511932611465454,
85
+ "logits/rejected": -2.4823100566864014,
86
+ "logps/chosen": -255.1762237548828,
87
+ "logps/rejected": -249.55859375,
88
+ "loss": 0.6856,
89
+ "rewards/accuracies": 0.5406249761581421,
90
+ "rewards/chosen": 0.0019304720917716622,
91
+ "rewards/margins": 0.0034704413264989853,
92
+ "rewards/rejected": -0.0015399687690660357,
93
+ "step": 50
94
+ },
95
+ {
96
+ "epoch": 0.06,
97
+ "learning_rate": 3.1249999999999997e-07,
98
+ "logits/chosen": -2.4945309162139893,
99
+ "logits/rejected": -2.5101318359375,
100
+ "logps/chosen": -288.4325256347656,
101
+ "logps/rejected": -261.18310546875,
102
+ "loss": 0.6867,
103
+ "rewards/accuracies": 0.503125011920929,
104
+ "rewards/chosen": -0.002425801707431674,
105
+ "rewards/margins": 0.002398231066763401,
106
+ "rewards/rejected": -0.004824032541364431,
107
+ "step": 60
108
+ },
109
+ {
110
+ "epoch": 0.07,
111
+ "learning_rate": 3.645833333333333e-07,
112
+ "logits/chosen": -2.521496534347534,
113
+ "logits/rejected": -2.506330966949463,
114
+ "logps/chosen": -275.6085510253906,
115
+ "logps/rejected": -259.58392333984375,
116
+ "loss": 0.6867,
117
+ "rewards/accuracies": 0.5375000238418579,
118
+ "rewards/chosen": 0.0025966810062527657,
119
+ "rewards/margins": 0.007796707563102245,
120
+ "rewards/rejected": -0.005200026091188192,
121
+ "step": 70
122
+ },
123
+ {
124
+ "epoch": 0.08,
125
+ "learning_rate": 4.1666666666666667e-07,
126
+ "logits/chosen": -2.456026315689087,
127
+ "logits/rejected": -2.4563703536987305,
128
+ "logps/chosen": -295.45941162109375,
129
+ "logps/rejected": -263.0380859375,
130
+ "loss": 0.6859,
131
+ "rewards/accuracies": 0.4593749940395355,
132
+ "rewards/chosen": -2.993037924170494e-05,
133
+ "rewards/margins": -0.000812772021163255,
134
+ "rewards/rejected": 0.0007828413508832455,
135
+ "step": 80
136
+ },
137
+ {
138
+ "epoch": 0.09,
139
+ "learning_rate": 4.6874999999999996e-07,
140
+ "logits/chosen": -2.4610087871551514,
141
+ "logits/rejected": -2.4633374214172363,
142
+ "logps/chosen": -282.82684326171875,
143
+ "logps/rejected": -245.01785278320312,
144
+ "loss": 0.6867,
145
+ "rewards/accuracies": 0.4906249940395355,
146
+ "rewards/chosen": -0.003978222608566284,
147
+ "rewards/margins": -0.003358404617756605,
148
+ "rewards/rejected": -0.0006198181654326618,
149
+ "step": 90
150
+ },
151
+ {
152
+ "epoch": 0.1,
153
+ "learning_rate": 4.976717112922002e-07,
154
+ "logits/chosen": -2.468822479248047,
155
+ "logits/rejected": -2.4451956748962402,
156
+ "logps/chosen": -264.9376525878906,
157
+ "logps/rejected": -259.81182861328125,
158
+ "loss": 0.6865,
159
+ "rewards/accuracies": 0.528124988079071,
160
+ "rewards/chosen": 0.004453591071069241,
161
+ "rewards/margins": 0.004475563298910856,
162
+ "rewards/rejected": -2.197279900428839e-05,
163
+ "step": 100
164
+ },
165
+ {
166
+ "epoch": 0.12,
167
+ "learning_rate": 4.918509895227007e-07,
168
+ "logits/chosen": -2.487288236618042,
169
+ "logits/rejected": -2.4807934761047363,
170
+ "logps/chosen": -263.60906982421875,
171
+ "logps/rejected": -244.46047973632812,
172
+ "loss": 0.6852,
173
+ "rewards/accuracies": 0.534375011920929,
174
+ "rewards/chosen": 0.00421065092086792,
175
+ "rewards/margins": 0.007397785782814026,
176
+ "rewards/rejected": -0.003187134861946106,
177
+ "step": 110
178
+ },
179
+ {
180
+ "epoch": 0.13,
181
+ "learning_rate": 4.860302677532014e-07,
182
+ "logits/chosen": -2.4685282707214355,
183
+ "logits/rejected": -2.497313976287842,
184
+ "logps/chosen": -277.75543212890625,
185
+ "logps/rejected": -245.1427764892578,
186
+ "loss": 0.6852,
187
+ "rewards/accuracies": 0.5531250238418579,
188
+ "rewards/chosen": 0.0019862186163663864,
189
+ "rewards/margins": 0.0036893251817673445,
190
+ "rewards/rejected": -0.0017031064489856362,
191
+ "step": 120
192
+ },
193
+ {
194
+ "epoch": 0.14,
195
+ "learning_rate": 4.802095459837019e-07,
196
+ "logits/chosen": -2.512012481689453,
197
+ "logits/rejected": -2.498793840408325,
198
+ "logps/chosen": -283.26959228515625,
199
+ "logps/rejected": -265.85369873046875,
200
+ "loss": 0.6838,
201
+ "rewards/accuracies": 0.534375011920929,
202
+ "rewards/chosen": 0.006660536862909794,
203
+ "rewards/margins": 0.007559786085039377,
204
+ "rewards/rejected": -0.0008992499788291752,
205
+ "step": 130
206
+ },
207
+ {
208
+ "epoch": 0.15,
209
+ "learning_rate": 4.743888242142026e-07,
210
+ "logits/chosen": -2.471496105194092,
211
+ "logits/rejected": -2.4844748973846436,
212
+ "logps/chosen": -255.3807373046875,
213
+ "logps/rejected": -242.02658081054688,
214
+ "loss": 0.6857,
215
+ "rewards/accuracies": 0.5062500238418579,
216
+ "rewards/chosen": 0.0008136004325933754,
217
+ "rewards/margins": 4.014703517896123e-05,
218
+ "rewards/rejected": 0.0007734533282928169,
219
+ "step": 140
220
+ },
221
+ {
222
+ "epoch": 0.16,
223
+ "learning_rate": 4.685681024447031e-07,
224
+ "logits/chosen": -2.471660614013672,
225
+ "logits/rejected": -2.463984966278076,
226
+ "logps/chosen": -287.92523193359375,
227
+ "logps/rejected": -260.02313232421875,
228
+ "loss": 0.6837,
229
+ "rewards/accuracies": 0.518750011920929,
230
+ "rewards/chosen": 0.00581919914111495,
231
+ "rewards/margins": 0.008211213164031506,
232
+ "rewards/rejected": -0.0023920147214084864,
233
+ "step": 150
234
+ },
235
+ {
236
+ "epoch": 0.17,
237
+ "learning_rate": 4.627473806752037e-07,
238
+ "logits/chosen": -2.520613193511963,
239
+ "logits/rejected": -2.4651389122009277,
240
+ "logps/chosen": -279.1468505859375,
241
+ "logps/rejected": -264.69915771484375,
242
+ "loss": 0.682,
243
+ "rewards/accuracies": 0.5531250238418579,
244
+ "rewards/chosen": 0.008319775573909283,
245
+ "rewards/margins": 0.011747308075428009,
246
+ "rewards/rejected": -0.003427532035857439,
247
+ "step": 160
248
+ },
249
+ {
250
+ "epoch": 0.18,
251
+ "learning_rate": 4.5692665890570433e-07,
252
+ "logits/chosen": -2.5029516220092773,
253
+ "logits/rejected": -2.496436595916748,
254
+ "logps/chosen": -279.58843994140625,
255
+ "logps/rejected": -243.44253540039062,
256
+ "loss": 0.6817,
257
+ "rewards/accuracies": 0.5874999761581421,
258
+ "rewards/chosen": 0.008326916955411434,
259
+ "rewards/margins": 0.014212583191692829,
260
+ "rewards/rejected": -0.00588566530495882,
261
+ "step": 170
262
+ },
263
+ {
264
+ "epoch": 0.19,
265
+ "learning_rate": 4.5110593713620486e-07,
266
+ "logits/chosen": -2.4697937965393066,
267
+ "logits/rejected": -2.4516141414642334,
268
+ "logps/chosen": -236.89675903320312,
269
+ "logps/rejected": -253.85598754882812,
270
+ "loss": 0.6832,
271
+ "rewards/accuracies": 0.5531250238418579,
272
+ "rewards/chosen": 0.00583054032176733,
273
+ "rewards/margins": 0.012070106342434883,
274
+ "rewards/rejected": -0.00623956648632884,
275
+ "step": 180
276
+ },
277
+ {
278
+ "epoch": 0.2,
279
+ "learning_rate": 4.4528521536670544e-07,
280
+ "logits/chosen": -2.5360610485076904,
281
+ "logits/rejected": -2.505577802658081,
282
+ "logps/chosen": -291.94586181640625,
283
+ "logps/rejected": -260.87054443359375,
284
+ "loss": 0.6801,
285
+ "rewards/accuracies": 0.637499988079071,
286
+ "rewards/chosen": 0.011800072155892849,
287
+ "rewards/margins": 0.02227787859737873,
288
+ "rewards/rejected": -0.010477803647518158,
289
+ "step": 190
290
+ },
291
+ {
292
+ "epoch": 0.21,
293
+ "learning_rate": 4.3946449359720607e-07,
294
+ "logits/chosen": -2.4782238006591797,
295
+ "logits/rejected": -2.508026599884033,
296
+ "logps/chosen": -258.73895263671875,
297
+ "logps/rejected": -252.0211639404297,
298
+ "loss": 0.6809,
299
+ "rewards/accuracies": 0.6031249761581421,
300
+ "rewards/chosen": 0.009132781066000462,
301
+ "rewards/margins": 0.0161266028881073,
302
+ "rewards/rejected": -0.006993822753429413,
303
+ "step": 200
304
+ },
305
+ {
306
+ "epoch": 0.22,
307
+ "learning_rate": 4.336437718277066e-07,
308
+ "logits/chosen": -2.4700803756713867,
309
+ "logits/rejected": -2.479269504547119,
310
+ "logps/chosen": -272.45904541015625,
311
+ "logps/rejected": -256.2876281738281,
312
+ "loss": 0.6783,
313
+ "rewards/accuracies": 0.574999988079071,
314
+ "rewards/chosen": 0.016153430566191673,
315
+ "rewards/margins": 0.023166943341493607,
316
+ "rewards/rejected": -0.007013511843979359,
317
+ "step": 210
318
+ },
319
+ {
320
+ "epoch": 0.23,
321
+ "learning_rate": 4.278230500582072e-07,
322
+ "logits/chosen": -2.4889426231384277,
323
+ "logits/rejected": -2.5127673149108887,
324
+ "logps/chosen": -267.21966552734375,
325
+ "logps/rejected": -258.29205322265625,
326
+ "loss": 0.6793,
327
+ "rewards/accuracies": 0.6031249761581421,
328
+ "rewards/chosen": 0.005786339286714792,
329
+ "rewards/margins": 0.01781514100730419,
330
+ "rewards/rejected": -0.012028800323605537,
331
+ "step": 220
332
+ },
333
+ {
334
+ "epoch": 0.24,
335
+ "learning_rate": 4.220023282887078e-07,
336
+ "logits/chosen": -2.4950485229492188,
337
+ "logits/rejected": -2.4980530738830566,
338
+ "logps/chosen": -295.85028076171875,
339
+ "logps/rejected": -259.36529541015625,
340
+ "loss": 0.6777,
341
+ "rewards/accuracies": 0.596875011920929,
342
+ "rewards/chosen": 0.007827522233128548,
343
+ "rewards/margins": 0.014727133326232433,
344
+ "rewards/rejected": -0.006899611093103886,
345
+ "step": 230
346
+ },
347
+ {
348
+ "epoch": 0.25,
349
+ "learning_rate": 4.1618160651920834e-07,
350
+ "logits/chosen": -2.466526508331299,
351
+ "logits/rejected": -2.4315407276153564,
352
+ "logps/chosen": -282.66192626953125,
353
+ "logps/rejected": -248.1639862060547,
354
+ "loss": 0.6778,
355
+ "rewards/accuracies": 0.5687500238418579,
356
+ "rewards/chosen": 0.01533576101064682,
357
+ "rewards/margins": 0.017889009788632393,
358
+ "rewards/rejected": -0.0025532490108162165,
359
+ "step": 240
360
+ },
361
+ {
362
+ "epoch": 0.26,
363
+ "learning_rate": 4.103608847497089e-07,
364
+ "logits/chosen": -2.4848718643188477,
365
+ "logits/rejected": -2.443408250808716,
366
+ "logps/chosen": -288.2550964355469,
367
+ "logps/rejected": -271.9901123046875,
368
+ "loss": 0.6761,
369
+ "rewards/accuracies": 0.621874988079071,
370
+ "rewards/chosen": 0.013339035212993622,
371
+ "rewards/margins": 0.023220960050821304,
372
+ "rewards/rejected": -0.009881924837827682,
373
+ "step": 250
374
+ },
375
+ {
376
+ "epoch": 0.27,
377
+ "learning_rate": 4.0454016298020956e-07,
378
+ "logits/chosen": -2.4762511253356934,
379
+ "logits/rejected": -2.446619749069214,
380
+ "logps/chosen": -268.5465393066406,
381
+ "logps/rejected": -251.9985809326172,
382
+ "loss": 0.6763,
383
+ "rewards/accuracies": 0.5625,
384
+ "rewards/chosen": 0.006427633110433817,
385
+ "rewards/margins": 0.015916740521788597,
386
+ "rewards/rejected": -0.009489107877016068,
387
+ "step": 260
388
+ },
389
+ {
390
+ "epoch": 0.28,
391
+ "learning_rate": 3.987194412107101e-07,
392
+ "logits/chosen": -2.4372754096984863,
393
+ "logits/rejected": -2.450491189956665,
394
+ "logps/chosen": -263.63262939453125,
395
+ "logps/rejected": -252.0638427734375,
396
+ "loss": 0.6735,
397
+ "rewards/accuracies": 0.6312500238418579,
398
+ "rewards/chosen": 0.013371949084103107,
399
+ "rewards/margins": 0.023976340889930725,
400
+ "rewards/rejected": -0.010604391805827618,
401
+ "step": 270
402
+ },
403
+ {
404
+ "epoch": 0.29,
405
+ "learning_rate": 3.9289871944121066e-07,
406
+ "logits/chosen": -2.5451080799102783,
407
+ "logits/rejected": -2.516885995864868,
408
+ "logps/chosen": -282.615234375,
409
+ "logps/rejected": -242.2822265625,
410
+ "loss": 0.6722,
411
+ "rewards/accuracies": 0.640625,
412
+ "rewards/chosen": 0.0192705187946558,
413
+ "rewards/margins": 0.029350727796554565,
414
+ "rewards/rejected": -0.010080209001898766,
415
+ "step": 280
416
+ },
417
+ {
418
+ "epoch": 0.3,
419
+ "learning_rate": 3.870779976717113e-07,
420
+ "logits/chosen": -2.4943687915802,
421
+ "logits/rejected": -2.5217864513397217,
422
+ "logps/chosen": -277.5215759277344,
423
+ "logps/rejected": -229.96298217773438,
424
+ "loss": 0.6722,
425
+ "rewards/accuracies": 0.6187499761581421,
426
+ "rewards/chosen": 0.015035443007946014,
427
+ "rewards/margins": 0.024382654577493668,
428
+ "rewards/rejected": -0.009347210638225079,
429
+ "step": 290
430
+ },
431
+ {
432
+ "epoch": 0.31,
433
+ "learning_rate": 3.812572759022118e-07,
434
+ "logits/chosen": -2.49491810798645,
435
+ "logits/rejected": -2.501307249069214,
436
+ "logps/chosen": -281.02484130859375,
437
+ "logps/rejected": -237.4232940673828,
438
+ "loss": 0.6726,
439
+ "rewards/accuracies": 0.628125011920929,
440
+ "rewards/chosen": 0.018051721155643463,
441
+ "rewards/margins": 0.032371118664741516,
442
+ "rewards/rejected": -0.014319395646452904,
443
+ "step": 300
444
+ },
445
+ {
446
+ "epoch": 0.32,
447
+ "learning_rate": 3.754365541327124e-07,
448
+ "logits/chosen": -2.4540202617645264,
449
+ "logits/rejected": -2.435542583465576,
450
+ "logps/chosen": -258.79193115234375,
451
+ "logps/rejected": -253.71694946289062,
452
+ "loss": 0.673,
453
+ "rewards/accuracies": 0.6468750238418579,
454
+ "rewards/chosen": 0.015593824908137321,
455
+ "rewards/margins": 0.0299003217369318,
456
+ "rewards/rejected": -0.014306495897471905,
457
+ "step": 310
458
+ },
459
+ {
460
+ "epoch": 0.33,
461
+ "learning_rate": 3.6961583236321304e-07,
462
+ "logits/chosen": -2.488478422164917,
463
+ "logits/rejected": -2.471566677093506,
464
+ "logps/chosen": -287.9583435058594,
465
+ "logps/rejected": -256.38427734375,
466
+ "loss": 0.6681,
467
+ "rewards/accuracies": 0.653124988079071,
468
+ "rewards/chosen": 0.01864878088235855,
469
+ "rewards/margins": 0.03612912446260452,
470
+ "rewards/rejected": -0.01748034544289112,
471
+ "step": 320
472
+ },
473
+ {
474
+ "epoch": 0.35,
475
+ "learning_rate": 3.637951105937136e-07,
476
+ "logits/chosen": -2.469587802886963,
477
+ "logits/rejected": -2.4525370597839355,
478
+ "logps/chosen": -249.15280151367188,
479
+ "logps/rejected": -235.3613739013672,
480
+ "loss": 0.6664,
481
+ "rewards/accuracies": 0.6187499761581421,
482
+ "rewards/chosen": 0.013197916559875011,
483
+ "rewards/margins": 0.030393391847610474,
484
+ "rewards/rejected": -0.01719547249376774,
485
+ "step": 330
486
+ },
487
+ {
488
+ "epoch": 0.36,
489
+ "learning_rate": 3.579743888242142e-07,
490
+ "logits/chosen": -2.486199378967285,
491
+ "logits/rejected": -2.491894006729126,
492
+ "logps/chosen": -279.8724060058594,
493
+ "logps/rejected": -249.26123046875,
494
+ "loss": 0.6691,
495
+ "rewards/accuracies": 0.59375,
496
+ "rewards/chosen": 0.016388490796089172,
497
+ "rewards/margins": 0.02631198987364769,
498
+ "rewards/rejected": -0.009923500940203667,
499
+ "step": 340
500
+ },
501
+ {
502
+ "epoch": 0.37,
503
+ "learning_rate": 3.521536670547148e-07,
504
+ "logits/chosen": -2.501917839050293,
505
+ "logits/rejected": -2.500124216079712,
506
+ "logps/chosen": -277.4295349121094,
507
+ "logps/rejected": -261.6769104003906,
508
+ "loss": 0.665,
509
+ "rewards/accuracies": 0.637499988079071,
510
+ "rewards/chosen": 0.019429894164204597,
511
+ "rewards/margins": 0.030850976705551147,
512
+ "rewards/rejected": -0.011421086266636848,
513
+ "step": 350
514
+ },
515
+ {
516
+ "epoch": 0.38,
517
+ "learning_rate": 3.4633294528521536e-07,
518
+ "logits/chosen": -2.5457377433776855,
519
+ "logits/rejected": -2.5352020263671875,
520
+ "logps/chosen": -280.8336486816406,
521
+ "logps/rejected": -259.36895751953125,
522
+ "loss": 0.6614,
523
+ "rewards/accuracies": 0.6156250238418579,
524
+ "rewards/chosen": 0.01814255118370056,
525
+ "rewards/margins": 0.03848281502723694,
526
+ "rewards/rejected": -0.02034026011824608,
527
+ "step": 360
528
+ },
529
+ {
530
+ "epoch": 0.39,
531
+ "learning_rate": 3.4051222351571594e-07,
532
+ "logits/chosen": -2.5180299282073975,
533
+ "logits/rejected": -2.4479191303253174,
534
+ "logps/chosen": -274.1128845214844,
535
+ "logps/rejected": -259.3582763671875,
536
+ "loss": 0.6664,
537
+ "rewards/accuracies": 0.643750011920929,
538
+ "rewards/chosen": 0.023864779621362686,
539
+ "rewards/margins": 0.041657593101263046,
540
+ "rewards/rejected": -0.01779281720519066,
541
+ "step": 370
542
+ },
543
+ {
544
+ "epoch": 0.4,
545
+ "learning_rate": 3.346915017462165e-07,
546
+ "logits/chosen": -2.504103899002075,
547
+ "logits/rejected": -2.4852070808410645,
548
+ "logps/chosen": -269.26409912109375,
549
+ "logps/rejected": -253.5343780517578,
550
+ "loss": 0.663,
551
+ "rewards/accuracies": 0.640625,
552
+ "rewards/chosen": 0.021114524453878403,
553
+ "rewards/margins": 0.042601972818374634,
554
+ "rewards/rejected": -0.021487446501851082,
555
+ "step": 380
556
+ },
557
+ {
558
+ "epoch": 0.41,
559
+ "learning_rate": 3.288707799767171e-07,
560
+ "logits/chosen": -2.4515504837036133,
561
+ "logits/rejected": -2.447246551513672,
562
+ "logps/chosen": -259.05401611328125,
563
+ "logps/rejected": -240.0911407470703,
564
+ "loss": 0.663,
565
+ "rewards/accuracies": 0.628125011920929,
566
+ "rewards/chosen": 0.014576256275177002,
567
+ "rewards/margins": 0.04004546254873276,
568
+ "rewards/rejected": -0.025469202548265457,
569
+ "step": 390
570
+ },
571
+ {
572
+ "epoch": 0.42,
573
+ "learning_rate": 3.230500582072177e-07,
574
+ "logits/chosen": -2.509997844696045,
575
+ "logits/rejected": -2.522238254547119,
576
+ "logps/chosen": -282.1490478515625,
577
+ "logps/rejected": -243.8563690185547,
578
+ "loss": 0.6624,
579
+ "rewards/accuracies": 0.637499988079071,
580
+ "rewards/chosen": 0.025715211406350136,
581
+ "rewards/margins": 0.043169617652893066,
582
+ "rewards/rejected": -0.01745440624654293,
583
+ "step": 400
584
+ },
585
+ {
586
+ "epoch": 0.43,
587
+ "learning_rate": 3.1722933643771827e-07,
588
+ "logits/chosen": -2.498192548751831,
589
+ "logits/rejected": -2.4921789169311523,
590
+ "logps/chosen": -271.2725830078125,
591
+ "logps/rejected": -244.8346710205078,
592
+ "loss": 0.6594,
593
+ "rewards/accuracies": 0.6468750238418579,
594
+ "rewards/chosen": 0.02634851261973381,
595
+ "rewards/margins": 0.0405191034078598,
596
+ "rewards/rejected": -0.014170585200190544,
597
+ "step": 410
598
+ },
599
+ {
600
+ "epoch": 0.44,
601
+ "learning_rate": 3.1140861466821885e-07,
602
+ "logits/chosen": -2.4811172485351562,
603
+ "logits/rejected": -2.4510648250579834,
604
+ "logps/chosen": -253.666015625,
605
+ "logps/rejected": -243.52877807617188,
606
+ "loss": 0.6617,
607
+ "rewards/accuracies": 0.628125011920929,
608
+ "rewards/chosen": 0.024721184745430946,
609
+ "rewards/margins": 0.03779374435544014,
610
+ "rewards/rejected": -0.013072559610009193,
611
+ "step": 420
612
+ },
613
+ {
614
+ "epoch": 0.45,
615
+ "learning_rate": 3.0558789289871943e-07,
616
+ "logits/chosen": -2.5075089931488037,
617
+ "logits/rejected": -2.4863505363464355,
618
+ "logps/chosen": -279.38818359375,
619
+ "logps/rejected": -254.1600799560547,
620
+ "loss": 0.6609,
621
+ "rewards/accuracies": 0.596875011920929,
622
+ "rewards/chosen": 0.02350917086005211,
623
+ "rewards/margins": 0.03294364735484123,
624
+ "rewards/rejected": -0.009434476494789124,
625
+ "step": 430
626
+ },
627
+ {
628
+ "epoch": 0.46,
629
+ "learning_rate": 2.9976717112922e-07,
630
+ "logits/chosen": -2.5201539993286133,
631
+ "logits/rejected": -2.511667013168335,
632
+ "logps/chosen": -272.9944152832031,
633
+ "logps/rejected": -233.72854614257812,
634
+ "loss": 0.6589,
635
+ "rewards/accuracies": 0.581250011920929,
636
+ "rewards/chosen": 0.020865267142653465,
637
+ "rewards/margins": 0.04071135073900223,
638
+ "rewards/rejected": -0.019846081733703613,
639
+ "step": 440
640
+ },
641
+ {
642
+ "epoch": 0.47,
643
+ "learning_rate": 2.939464493597206e-07,
644
+ "logits/chosen": -2.4186248779296875,
645
+ "logits/rejected": -2.3882439136505127,
646
+ "logps/chosen": -244.801025390625,
647
+ "logps/rejected": -241.0476837158203,
648
+ "loss": 0.6588,
649
+ "rewards/accuracies": 0.637499988079071,
650
+ "rewards/chosen": 0.019629117101430893,
651
+ "rewards/margins": 0.037906430661678314,
652
+ "rewards/rejected": -0.01827731356024742,
653
+ "step": 450
654
+ },
655
+ {
656
+ "epoch": 0.48,
657
+ "learning_rate": 2.8812572759022117e-07,
658
+ "logits/chosen": -2.5232837200164795,
659
+ "logits/rejected": -2.5000243186950684,
660
+ "logps/chosen": -294.4573669433594,
661
+ "logps/rejected": -256.6082458496094,
662
+ "loss": 0.6559,
663
+ "rewards/accuracies": 0.6156250238418579,
664
+ "rewards/chosen": 0.02450462244451046,
665
+ "rewards/margins": 0.0395827516913414,
666
+ "rewards/rejected": -0.015078130178153515,
667
+ "step": 460
668
+ },
669
+ {
670
+ "epoch": 0.49,
671
+ "learning_rate": 2.8230500582072175e-07,
672
+ "logits/chosen": -2.4724531173706055,
673
+ "logits/rejected": -2.4754977226257324,
674
+ "logps/chosen": -283.3002624511719,
675
+ "logps/rejected": -251.4059295654297,
676
+ "loss": 0.6557,
677
+ "rewards/accuracies": 0.65625,
678
+ "rewards/chosen": 0.024192675948143005,
679
+ "rewards/margins": 0.05236636474728584,
680
+ "rewards/rejected": -0.028173688799142838,
681
+ "step": 470
682
+ },
683
+ {
684
+ "epoch": 0.5,
685
+ "learning_rate": 2.7648428405122233e-07,
686
+ "logits/chosen": -2.410810947418213,
687
+ "logits/rejected": -2.4008870124816895,
688
+ "logps/chosen": -242.271484375,
689
+ "logps/rejected": -250.8043212890625,
690
+ "loss": 0.6593,
691
+ "rewards/accuracies": 0.6031249761581421,
692
+ "rewards/chosen": 0.012235969305038452,
693
+ "rewards/margins": 0.02636186219751835,
694
+ "rewards/rejected": -0.01412589568644762,
695
+ "step": 480
696
+ },
697
+ {
698
+ "epoch": 0.51,
699
+ "learning_rate": 2.706635622817229e-07,
700
+ "logits/chosen": -2.4645464420318604,
701
+ "logits/rejected": -2.469316244125366,
702
+ "logps/chosen": -272.09796142578125,
703
+ "logps/rejected": -238.6561737060547,
704
+ "loss": 0.6533,
705
+ "rewards/accuracies": 0.671875,
706
+ "rewards/chosen": 0.024095356464385986,
707
+ "rewards/margins": 0.05057717487215996,
708
+ "rewards/rejected": -0.02648181840777397,
709
+ "step": 490
710
+ },
711
+ {
712
+ "epoch": 0.52,
713
+ "learning_rate": 2.648428405122235e-07,
714
+ "logits/chosen": -2.508164167404175,
715
+ "logits/rejected": -2.4578819274902344,
716
+ "logps/chosen": -277.1492004394531,
717
+ "logps/rejected": -256.0687561035156,
718
+ "loss": 0.6531,
719
+ "rewards/accuracies": 0.640625,
720
+ "rewards/chosen": 0.03192506358027458,
721
+ "rewards/margins": 0.05114533379673958,
722
+ "rewards/rejected": -0.019220268353819847,
723
+ "step": 500
724
+ },
725
+ {
726
+ "epoch": 0.53,
727
+ "learning_rate": 2.590221187427241e-07,
728
+ "logits/chosen": -2.5105607509613037,
729
+ "logits/rejected": -2.4753470420837402,
730
+ "logps/chosen": -263.40966796875,
731
+ "logps/rejected": -244.8257598876953,
732
+ "loss": 0.6522,
733
+ "rewards/accuracies": 0.625,
734
+ "rewards/chosen": 0.027492288500070572,
735
+ "rewards/margins": 0.04991314187645912,
736
+ "rewards/rejected": -0.022420858964323997,
737
+ "step": 510
738
+ },
739
+ {
740
+ "epoch": 0.54,
741
+ "learning_rate": 2.5320139697322466e-07,
742
+ "logits/chosen": -2.5335867404937744,
743
+ "logits/rejected": -2.468331813812256,
744
+ "logps/chosen": -272.64385986328125,
745
+ "logps/rejected": -244.65286254882812,
746
+ "loss": 0.651,
747
+ "rewards/accuracies": 0.65625,
748
+ "rewards/chosen": 0.03846021741628647,
749
+ "rewards/margins": 0.06949218362569809,
750
+ "rewards/rejected": -0.03103196993470192,
751
+ "step": 520
752
+ },
753
+ {
754
+ "epoch": 0.55,
755
+ "learning_rate": 2.4738067520372524e-07,
756
+ "logits/chosen": -2.505664110183716,
757
+ "logits/rejected": -2.512172222137451,
758
+ "logps/chosen": -280.70819091796875,
759
+ "logps/rejected": -255.12228393554688,
760
+ "loss": 0.6487,
761
+ "rewards/accuracies": 0.668749988079071,
762
+ "rewards/chosen": 0.035160940140485764,
763
+ "rewards/margins": 0.06470004469156265,
764
+ "rewards/rejected": -0.029539108276367188,
765
+ "step": 530
766
+ },
767
+ {
768
+ "epoch": 0.57,
769
+ "learning_rate": 2.415599534342258e-07,
770
+ "logits/chosen": -2.5341105461120605,
771
+ "logits/rejected": -2.4813685417175293,
772
+ "logps/chosen": -280.51751708984375,
773
+ "logps/rejected": -252.9778289794922,
774
+ "loss": 0.6471,
775
+ "rewards/accuracies": 0.653124988079071,
776
+ "rewards/chosen": 0.03973756358027458,
777
+ "rewards/margins": 0.06431148201227188,
778
+ "rewards/rejected": -0.02457391656935215,
779
+ "step": 540
780
+ },
781
+ {
782
+ "epoch": 0.58,
783
+ "learning_rate": 2.3573923166472642e-07,
784
+ "logits/chosen": -2.506739377975464,
785
+ "logits/rejected": -2.487623691558838,
786
+ "logps/chosen": -272.10675048828125,
787
+ "logps/rejected": -258.69647216796875,
788
+ "loss": 0.6479,
789
+ "rewards/accuracies": 0.6343749761581421,
790
+ "rewards/chosen": 0.028482938185334206,
791
+ "rewards/margins": 0.057554639875888824,
792
+ "rewards/rejected": -0.02907169796526432,
793
+ "step": 550
794
+ },
795
+ {
796
+ "epoch": 0.59,
797
+ "learning_rate": 2.2991850989522698e-07,
798
+ "logits/chosen": -2.442371129989624,
799
+ "logits/rejected": -2.4590439796447754,
800
+ "logps/chosen": -280.783935546875,
801
+ "logps/rejected": -227.5161895751953,
802
+ "loss": 0.6513,
803
+ "rewards/accuracies": 0.628125011920929,
804
+ "rewards/chosen": 0.02295442670583725,
805
+ "rewards/margins": 0.05334927886724472,
806
+ "rewards/rejected": -0.03039485774934292,
807
+ "step": 560
808
+ },
809
+ {
810
+ "epoch": 0.6,
811
+ "learning_rate": 2.2409778812572759e-07,
812
+ "logits/chosen": -2.470194101333618,
813
+ "logits/rejected": -2.441986560821533,
814
+ "logps/chosen": -255.4265594482422,
815
+ "logps/rejected": -241.7792205810547,
816
+ "loss": 0.6487,
817
+ "rewards/accuracies": 0.6343749761581421,
818
+ "rewards/chosen": 0.024136796593666077,
819
+ "rewards/margins": 0.04882895201444626,
820
+ "rewards/rejected": -0.02469216100871563,
821
+ "step": 570
822
+ },
823
+ {
824
+ "epoch": 0.61,
825
+ "learning_rate": 2.1827706635622817e-07,
826
+ "logits/chosen": -2.5127058029174805,
827
+ "logits/rejected": -2.496788263320923,
828
+ "logps/chosen": -262.003662109375,
829
+ "logps/rejected": -251.4923858642578,
830
+ "loss": 0.6508,
831
+ "rewards/accuracies": 0.6937500238418579,
832
+ "rewards/chosen": 0.025721842423081398,
833
+ "rewards/margins": 0.056790102273225784,
834
+ "rewards/rejected": -0.031068259850144386,
835
+ "step": 580
836
+ },
837
+ {
838
+ "epoch": 0.62,
839
+ "learning_rate": 2.1245634458672875e-07,
840
+ "logits/chosen": -2.4597954750061035,
841
+ "logits/rejected": -2.4486401081085205,
842
+ "logps/chosen": -272.46258544921875,
843
+ "logps/rejected": -250.1220703125,
844
+ "loss": 0.6465,
845
+ "rewards/accuracies": 0.671875,
846
+ "rewards/chosen": 0.03377040475606918,
847
+ "rewards/margins": 0.05980812385678291,
848
+ "rewards/rejected": -0.02603771723806858,
849
+ "step": 590
850
+ },
851
+ {
852
+ "epoch": 0.63,
853
+ "learning_rate": 2.0663562281722933e-07,
854
+ "logits/chosen": -2.468313217163086,
855
+ "logits/rejected": -2.4899404048919678,
856
+ "logps/chosen": -280.9910888671875,
857
+ "logps/rejected": -279.2497863769531,
858
+ "loss": 0.6493,
859
+ "rewards/accuracies": 0.6468750238418579,
860
+ "rewards/chosen": 0.03328691050410271,
861
+ "rewards/margins": 0.05028299614787102,
862
+ "rewards/rejected": -0.016996093094348907,
863
+ "step": 600
864
+ },
865
+ {
866
+ "epoch": 0.64,
867
+ "learning_rate": 2.008149010477299e-07,
868
+ "logits/chosen": -2.515079975128174,
869
+ "logits/rejected": -2.471588611602783,
870
+ "logps/chosen": -284.3990478515625,
871
+ "logps/rejected": -246.21835327148438,
872
+ "loss": 0.6456,
873
+ "rewards/accuracies": 0.637499988079071,
874
+ "rewards/chosen": 0.03540956228971481,
875
+ "rewards/margins": 0.062371380627155304,
876
+ "rewards/rejected": -0.026961814612150192,
877
+ "step": 610
878
+ },
879
+ {
880
+ "epoch": 0.65,
881
+ "learning_rate": 1.949941792782305e-07,
882
+ "logits/chosen": -2.5138742923736572,
883
+ "logits/rejected": -2.451676845550537,
884
+ "logps/chosen": -272.3465881347656,
885
+ "logps/rejected": -240.0901641845703,
886
+ "loss": 0.6492,
887
+ "rewards/accuracies": 0.6343749761581421,
888
+ "rewards/chosen": 0.03495832532644272,
889
+ "rewards/margins": 0.05197330191731453,
890
+ "rewards/rejected": -0.017014967277646065,
891
+ "step": 620
892
+ },
893
+ {
894
+ "epoch": 0.66,
895
+ "learning_rate": 1.8917345750873107e-07,
896
+ "logits/chosen": -2.4583332538604736,
897
+ "logits/rejected": -2.468278169631958,
898
+ "logps/chosen": -264.42913818359375,
899
+ "logps/rejected": -256.8759765625,
900
+ "loss": 0.6451,
901
+ "rewards/accuracies": 0.637499988079071,
902
+ "rewards/chosen": 0.02500128373503685,
903
+ "rewards/margins": 0.04210829734802246,
904
+ "rewards/rejected": -0.01710701361298561,
905
+ "step": 630
906
+ },
907
+ {
908
+ "epoch": 0.67,
909
+ "learning_rate": 1.8335273573923165e-07,
910
+ "logits/chosen": -2.5137476921081543,
911
+ "logits/rejected": -2.473841667175293,
912
+ "logps/chosen": -252.57705688476562,
913
+ "logps/rejected": -230.48471069335938,
914
+ "loss": 0.6434,
915
+ "rewards/accuracies": 0.640625,
916
+ "rewards/chosen": 0.03589923307299614,
917
+ "rewards/margins": 0.06454572081565857,
918
+ "rewards/rejected": -0.028646480292081833,
919
+ "step": 640
920
+ },
921
+ {
922
+ "epoch": 0.68,
923
+ "learning_rate": 1.7753201396973226e-07,
924
+ "logits/chosen": -2.45261549949646,
925
+ "logits/rejected": -2.4653515815734863,
926
+ "logps/chosen": -274.7919006347656,
927
+ "logps/rejected": -261.8819580078125,
928
+ "loss": 0.6407,
929
+ "rewards/accuracies": 0.6468750238418579,
930
+ "rewards/chosen": 0.03197002038359642,
931
+ "rewards/margins": 0.07201725244522095,
932
+ "rewards/rejected": -0.040047239512205124,
933
+ "step": 650
934
+ },
935
+ {
936
+ "epoch": 0.69,
937
+ "learning_rate": 1.7171129220023281e-07,
938
+ "logits/chosen": -2.4911041259765625,
939
+ "logits/rejected": -2.4684813022613525,
940
+ "logps/chosen": -267.3164978027344,
941
+ "logps/rejected": -232.7432403564453,
942
+ "loss": 0.6489,
943
+ "rewards/accuracies": 0.6499999761581421,
944
+ "rewards/chosen": 0.022824671119451523,
945
+ "rewards/margins": 0.04568014293909073,
946
+ "rewards/rejected": -0.022855471819639206,
947
+ "step": 660
948
+ },
949
+ {
950
+ "epoch": 0.7,
951
+ "learning_rate": 1.658905704307334e-07,
952
+ "logits/chosen": -2.4661662578582764,
953
+ "logits/rejected": -2.429795026779175,
954
+ "logps/chosen": -267.6933898925781,
955
+ "logps/rejected": -238.0891876220703,
956
+ "loss": 0.6461,
957
+ "rewards/accuracies": 0.65625,
958
+ "rewards/chosen": 0.029722299426794052,
959
+ "rewards/margins": 0.061273299157619476,
960
+ "rewards/rejected": -0.031550996005535126,
961
+ "step": 670
962
+ },
963
+ {
964
+ "epoch": 0.71,
965
+ "learning_rate": 1.60069848661234e-07,
966
+ "logits/chosen": -2.494511127471924,
967
+ "logits/rejected": -2.505267381668091,
968
+ "logps/chosen": -287.6944274902344,
969
+ "logps/rejected": -259.6459045410156,
970
+ "loss": 0.6466,
971
+ "rewards/accuracies": 0.640625,
972
+ "rewards/chosen": 0.041329581290483475,
973
+ "rewards/margins": 0.05584716796875,
974
+ "rewards/rejected": -0.014517592266201973,
975
+ "step": 680
976
+ },
977
+ {
978
+ "epoch": 0.72,
979
+ "learning_rate": 1.5424912689173456e-07,
980
+ "logits/chosen": -2.4804582595825195,
981
+ "logits/rejected": -2.462634325027466,
982
+ "logps/chosen": -270.22698974609375,
983
+ "logps/rejected": -227.7113800048828,
984
+ "loss": 0.6407,
985
+ "rewards/accuracies": 0.671875,
986
+ "rewards/chosen": 0.037567246705293655,
987
+ "rewards/margins": 0.0671004056930542,
988
+ "rewards/rejected": -0.02953316643834114,
989
+ "step": 690
990
+ },
991
+ {
992
+ "epoch": 0.73,
993
+ "learning_rate": 1.4842840512223514e-07,
994
+ "logits/chosen": -2.468341112136841,
995
+ "logits/rejected": -2.4682674407958984,
996
+ "logps/chosen": -257.55096435546875,
997
+ "logps/rejected": -232.13906860351562,
998
+ "loss": 0.647,
999
+ "rewards/accuracies": 0.671875,
1000
+ "rewards/chosen": 0.03749538213014603,
1001
+ "rewards/margins": 0.0758819431066513,
1002
+ "rewards/rejected": -0.03838656097650528,
1003
+ "step": 700
1004
+ },
1005
+ {
1006
+ "epoch": 0.74,
1007
+ "learning_rate": 1.4260768335273574e-07,
1008
+ "logits/chosen": -2.5157179832458496,
1009
+ "logits/rejected": -2.497849702835083,
1010
+ "logps/chosen": -264.9339904785156,
1011
+ "logps/rejected": -229.9503936767578,
1012
+ "loss": 0.6375,
1013
+ "rewards/accuracies": 0.6312500238418579,
1014
+ "rewards/chosen": 0.03981786221265793,
1015
+ "rewards/margins": 0.06821642816066742,
1016
+ "rewards/rejected": -0.02839856967329979,
1017
+ "step": 710
1018
+ },
1019
+ {
1020
+ "epoch": 0.75,
1021
+ "learning_rate": 1.3678696158323632e-07,
1022
+ "logits/chosen": -2.4586503505706787,
1023
+ "logits/rejected": -2.4603028297424316,
1024
+ "logps/chosen": -262.70880126953125,
1025
+ "logps/rejected": -237.71212768554688,
1026
+ "loss": 0.645,
1027
+ "rewards/accuracies": 0.6312500238418579,
1028
+ "rewards/chosen": 0.025259777903556824,
1029
+ "rewards/margins": 0.0490126870572567,
1030
+ "rewards/rejected": -0.023752911016345024,
1031
+ "step": 720
1032
+ },
1033
+ {
1034
+ "epoch": 0.76,
1035
+ "learning_rate": 1.3096623981373688e-07,
1036
+ "logits/chosen": -2.502094268798828,
1037
+ "logits/rejected": -2.4800729751586914,
1038
+ "logps/chosen": -282.3802185058594,
1039
+ "logps/rejected": -236.8638153076172,
1040
+ "loss": 0.6438,
1041
+ "rewards/accuracies": 0.6656249761581421,
1042
+ "rewards/chosen": 0.04312821850180626,
1043
+ "rewards/margins": 0.07856440544128418,
1044
+ "rewards/rejected": -0.03543618693947792,
1045
+ "step": 730
1046
+ },
1047
+ {
1048
+ "epoch": 0.77,
1049
+ "learning_rate": 1.2514551804423749e-07,
1050
+ "logits/chosen": -2.4671387672424316,
1051
+ "logits/rejected": -2.514324903488159,
1052
+ "logps/chosen": -285.3311462402344,
1053
+ "logps/rejected": -263.767822265625,
1054
+ "loss": 0.642,
1055
+ "rewards/accuracies": 0.684374988079071,
1056
+ "rewards/chosen": 0.04080774262547493,
1057
+ "rewards/margins": 0.07315204292535782,
1058
+ "rewards/rejected": -0.03234430402517319,
1059
+ "step": 740
1060
+ },
1061
+ {
1062
+ "epoch": 0.79,
1063
+ "learning_rate": 1.1932479627473807e-07,
1064
+ "logits/chosen": -2.490058183670044,
1065
+ "logits/rejected": -2.4621422290802,
1066
+ "logps/chosen": -251.80148315429688,
1067
+ "logps/rejected": -249.9593048095703,
1068
+ "loss": 0.6395,
1069
+ "rewards/accuracies": 0.6625000238418579,
1070
+ "rewards/chosen": 0.028191978111863136,
1071
+ "rewards/margins": 0.06258732825517654,
1072
+ "rewards/rejected": -0.03439534455537796,
1073
+ "step": 750
1074
+ },
1075
+ {
1076
+ "epoch": 0.8,
1077
+ "learning_rate": 1.1350407450523865e-07,
1078
+ "logits/chosen": -2.470336437225342,
1079
+ "logits/rejected": -2.5136280059814453,
1080
+ "logps/chosen": -281.42767333984375,
1081
+ "logps/rejected": -254.0972137451172,
1082
+ "loss": 0.6419,
1083
+ "rewards/accuracies": 0.637499988079071,
1084
+ "rewards/chosen": 0.036293573677539825,
1085
+ "rewards/margins": 0.06266774237155914,
1086
+ "rewards/rejected": -0.02637416496872902,
1087
+ "step": 760
1088
+ },
1089
+ {
1090
+ "epoch": 0.81,
1091
+ "learning_rate": 1.0768335273573923e-07,
1092
+ "logits/chosen": -2.5531864166259766,
1093
+ "logits/rejected": -2.545531749725342,
1094
+ "logps/chosen": -305.70111083984375,
1095
+ "logps/rejected": -276.01019287109375,
1096
+ "loss": 0.64,
1097
+ "rewards/accuracies": 0.621874988079071,
1098
+ "rewards/chosen": 0.046415045857429504,
1099
+ "rewards/margins": 0.07657970488071442,
1100
+ "rewards/rejected": -0.030164653435349464,
1101
+ "step": 770
1102
+ },
1103
+ {
1104
+ "epoch": 0.82,
1105
+ "learning_rate": 1.0186263096623981e-07,
1106
+ "logits/chosen": -2.484541416168213,
1107
+ "logits/rejected": -2.4959442615509033,
1108
+ "logps/chosen": -272.62445068359375,
1109
+ "logps/rejected": -249.4247589111328,
1110
+ "loss": 0.6416,
1111
+ "rewards/accuracies": 0.6187499761581421,
1112
+ "rewards/chosen": 0.03473570942878723,
1113
+ "rewards/margins": 0.057318903505802155,
1114
+ "rewards/rejected": -0.022583190351724625,
1115
+ "step": 780
1116
+ },
1117
+ {
1118
+ "epoch": 0.83,
1119
+ "learning_rate": 9.604190919674039e-08,
1120
+ "logits/chosen": -2.480217218399048,
1121
+ "logits/rejected": -2.4997220039367676,
1122
+ "logps/chosen": -253.22061157226562,
1123
+ "logps/rejected": -248.5176239013672,
1124
+ "loss": 0.6447,
1125
+ "rewards/accuracies": 0.612500011920929,
1126
+ "rewards/chosen": 0.02552936039865017,
1127
+ "rewards/margins": 0.05032141134142876,
1128
+ "rewards/rejected": -0.024792049080133438,
1129
+ "step": 790
1130
+ },
1131
+ {
1132
+ "epoch": 0.84,
1133
+ "learning_rate": 9.022118742724097e-08,
1134
+ "logits/chosen": -2.4872233867645264,
1135
+ "logits/rejected": -2.461895704269409,
1136
+ "logps/chosen": -271.40191650390625,
1137
+ "logps/rejected": -261.29241943359375,
1138
+ "loss": 0.6352,
1139
+ "rewards/accuracies": 0.640625,
1140
+ "rewards/chosen": 0.035524625331163406,
1141
+ "rewards/margins": 0.0665384978055954,
1142
+ "rewards/rejected": -0.031013870611786842,
1143
+ "step": 800
1144
+ },
1145
+ {
1146
+ "epoch": 0.85,
1147
+ "learning_rate": 8.440046565774157e-08,
1148
+ "logits/chosen": -2.462954521179199,
1149
+ "logits/rejected": -2.472172498703003,
1150
+ "logps/chosen": -272.635986328125,
1151
+ "logps/rejected": -229.35366821289062,
1152
+ "loss": 0.6432,
1153
+ "rewards/accuracies": 0.65625,
1154
+ "rewards/chosen": 0.03845527023077011,
1155
+ "rewards/margins": 0.05895563215017319,
1156
+ "rewards/rejected": -0.020500360056757927,
1157
+ "step": 810
1158
+ },
1159
+ {
1160
+ "epoch": 0.86,
1161
+ "learning_rate": 7.857974388824213e-08,
1162
+ "logits/chosen": -2.481274127960205,
1163
+ "logits/rejected": -2.4623420238494873,
1164
+ "logps/chosen": -282.552734375,
1165
+ "logps/rejected": -246.048095703125,
1166
+ "loss": 0.6457,
1167
+ "rewards/accuracies": 0.6312500238418579,
1168
+ "rewards/chosen": 0.04651721939444542,
1169
+ "rewards/margins": 0.07211866974830627,
1170
+ "rewards/rejected": -0.025601446628570557,
1171
+ "step": 820
1172
+ },
1173
+ {
1174
+ "epoch": 0.87,
1175
+ "learning_rate": 7.275902211874273e-08,
1176
+ "logits/chosen": -2.495702028274536,
1177
+ "logits/rejected": -2.4765870571136475,
1178
+ "logps/chosen": -249.3843536376953,
1179
+ "logps/rejected": -278.4346923828125,
1180
+ "loss": 0.644,
1181
+ "rewards/accuracies": 0.684374988079071,
1182
+ "rewards/chosen": 0.021697301417589188,
1183
+ "rewards/margins": 0.059917084872722626,
1184
+ "rewards/rejected": -0.038219790905714035,
1185
+ "step": 830
1186
+ },
1187
+ {
1188
+ "epoch": 0.88,
1189
+ "learning_rate": 6.693830034924331e-08,
1190
+ "logits/chosen": -2.4697625637054443,
1191
+ "logits/rejected": -2.4722721576690674,
1192
+ "logps/chosen": -265.5938415527344,
1193
+ "logps/rejected": -257.2969665527344,
1194
+ "loss": 0.6376,
1195
+ "rewards/accuracies": 0.653124988079071,
1196
+ "rewards/chosen": 0.028097212314605713,
1197
+ "rewards/margins": 0.046820152550935745,
1198
+ "rewards/rejected": -0.018722938373684883,
1199
+ "step": 840
1200
+ },
1201
+ {
1202
+ "epoch": 0.89,
1203
+ "learning_rate": 6.111757857974389e-08,
1204
+ "logits/chosen": -2.4835047721862793,
1205
+ "logits/rejected": -2.5235934257507324,
1206
+ "logps/chosen": -263.8602294921875,
1207
+ "logps/rejected": -241.2860870361328,
1208
+ "loss": 0.6443,
1209
+ "rewards/accuracies": 0.640625,
1210
+ "rewards/chosen": 0.03554076701402664,
1211
+ "rewards/margins": 0.065843865275383,
1212
+ "rewards/rejected": -0.030303100124001503,
1213
+ "step": 850
1214
+ },
1215
+ {
1216
+ "epoch": 0.9,
1217
+ "learning_rate": 5.529685681024446e-08,
1218
+ "logits/chosen": -2.446396589279175,
1219
+ "logits/rejected": -2.447554111480713,
1220
+ "logps/chosen": -278.2039489746094,
1221
+ "logps/rejected": -267.393798828125,
1222
+ "loss": 0.6433,
1223
+ "rewards/accuracies": 0.6312500238418579,
1224
+ "rewards/chosen": 0.02485939860343933,
1225
+ "rewards/margins": 0.04789874702692032,
1226
+ "rewards/rejected": -0.023039352148771286,
1227
+ "step": 860
1228
+ },
1229
+ {
1230
+ "epoch": 0.91,
1231
+ "learning_rate": 4.947613504074505e-08,
1232
+ "logits/chosen": -2.451906442642212,
1233
+ "logits/rejected": -2.4401917457580566,
1234
+ "logps/chosen": -254.4597625732422,
1235
+ "logps/rejected": -240.5189666748047,
1236
+ "loss": 0.6406,
1237
+ "rewards/accuracies": 0.596875011920929,
1238
+ "rewards/chosen": 0.03511255979537964,
1239
+ "rewards/margins": 0.05534617230296135,
1240
+ "rewards/rejected": -0.020233619958162308,
1241
+ "step": 870
1242
+ },
1243
+ {
1244
+ "epoch": 0.92,
1245
+ "learning_rate": 4.365541327124563e-08,
1246
+ "logits/chosen": -2.507020950317383,
1247
+ "logits/rejected": -2.4901695251464844,
1248
+ "logps/chosen": -260.8884582519531,
1249
+ "logps/rejected": -251.20938110351562,
1250
+ "loss": 0.6417,
1251
+ "rewards/accuracies": 0.6312500238418579,
1252
+ "rewards/chosen": 0.04014205187559128,
1253
+ "rewards/margins": 0.060112785547971725,
1254
+ "rewards/rejected": -0.01997072994709015,
1255
+ "step": 880
1256
+ },
1257
+ {
1258
+ "epoch": 0.93,
1259
+ "learning_rate": 3.783469150174622e-08,
1260
+ "logits/chosen": -2.5302183628082275,
1261
+ "logits/rejected": -2.5431013107299805,
1262
+ "logps/chosen": -254.4590606689453,
1263
+ "logps/rejected": -259.6826477050781,
1264
+ "loss": 0.6409,
1265
+ "rewards/accuracies": 0.643750011920929,
1266
+ "rewards/chosen": 0.02364221215248108,
1267
+ "rewards/margins": 0.04719501733779907,
1268
+ "rewards/rejected": -0.023552805185317993,
1269
+ "step": 890
1270
+ },
1271
+ {
1272
+ "epoch": 0.94,
1273
+ "learning_rate": 3.20139697322468e-08,
1274
+ "logits/chosen": -2.4807305335998535,
1275
+ "logits/rejected": -2.5058140754699707,
1276
+ "logps/chosen": -264.16021728515625,
1277
+ "logps/rejected": -249.19570922851562,
1278
+ "loss": 0.6452,
1279
+ "rewards/accuracies": 0.628125011920929,
1280
+ "rewards/chosen": 0.029653768986463547,
1281
+ "rewards/margins": 0.05728424713015556,
1282
+ "rewards/rejected": -0.027630474418401718,
1283
+ "step": 900
1284
+ },
1285
+ {
1286
+ "epoch": 0.95,
1287
+ "learning_rate": 2.619324796274738e-08,
1288
+ "logits/chosen": -2.48866605758667,
1289
+ "logits/rejected": -2.5024585723876953,
1290
+ "logps/chosen": -289.67401123046875,
1291
+ "logps/rejected": -261.8035583496094,
1292
+ "loss": 0.6424,
1293
+ "rewards/accuracies": 0.643750011920929,
1294
+ "rewards/chosen": 0.041562773287296295,
1295
+ "rewards/margins": 0.06672655045986176,
1296
+ "rewards/rejected": -0.025163773447275162,
1297
+ "step": 910
1298
+ },
1299
+ {
1300
+ "epoch": 0.96,
1301
+ "learning_rate": 2.037252619324796e-08,
1302
+ "logits/chosen": -2.4807181358337402,
1303
+ "logits/rejected": -2.4651541709899902,
1304
+ "logps/chosen": -270.3717346191406,
1305
+ "logps/rejected": -248.5556182861328,
1306
+ "loss": 0.6415,
1307
+ "rewards/accuracies": 0.653124988079071,
1308
+ "rewards/chosen": 0.022484585642814636,
1309
+ "rewards/margins": 0.05588115006685257,
1310
+ "rewards/rejected": -0.033396560698747635,
1311
+ "step": 920
1312
+ },
1313
+ {
1314
+ "epoch": 0.97,
1315
+ "learning_rate": 1.4551804423748545e-08,
1316
+ "logits/chosen": -2.4763240814208984,
1317
+ "logits/rejected": -2.4676055908203125,
1318
+ "logps/chosen": -272.7355041503906,
1319
+ "logps/rejected": -254.0940704345703,
1320
+ "loss": 0.6432,
1321
+ "rewards/accuracies": 0.65625,
1322
+ "rewards/chosen": 0.046184636652469635,
1323
+ "rewards/margins": 0.0702408105134964,
1324
+ "rewards/rejected": -0.02405618131160736,
1325
+ "step": 930
1326
+ },
1327
+ {
1328
+ "epoch": 0.98,
1329
+ "learning_rate": 8.731082654249125e-09,
1330
+ "logits/chosen": -2.4857518672943115,
1331
+ "logits/rejected": -2.5030598640441895,
1332
+ "logps/chosen": -275.62445068359375,
1333
+ "logps/rejected": -264.8072509765625,
1334
+ "loss": 0.6407,
1335
+ "rewards/accuracies": 0.671875,
1336
+ "rewards/chosen": 0.036477264016866684,
1337
+ "rewards/margins": 0.06461935490369797,
1338
+ "rewards/rejected": -0.02814210020005703,
1339
+ "step": 940
1340
+ },
1341
+ {
1342
+ "epoch": 0.99,
1343
+ "learning_rate": 2.910360884749709e-09,
1344
+ "logits/chosen": -2.4887900352478027,
1345
+ "logits/rejected": -2.494776725769043,
1346
+ "logps/chosen": -259.7311706542969,
1347
+ "logps/rejected": -253.02346801757812,
1348
+ "loss": 0.6442,
1349
+ "rewards/accuracies": 0.65625,
1350
+ "rewards/chosen": 0.02119762822985649,
1351
+ "rewards/margins": 0.04852701723575592,
1352
+ "rewards/rejected": -0.02732938900589943,
1353
+ "step": 950
1354
+ },
1355
+ {
1356
+ "epoch": 1.0,
1357
+ "eval_logits/chosen": -2.4944536685943604,
1358
+ "eval_logits/rejected": -2.4963433742523193,
1359
+ "eval_logps/chosen": -269.25555419921875,
1360
+ "eval_logps/rejected": -253.21238708496094,
1361
+ "eval_loss": 0.6399702429771423,
1362
+ "eval_rewards/accuracies": 0.6370000243186951,
1363
+ "eval_rewards/chosen": 0.030110126361250877,
1364
+ "eval_rewards/margins": 0.05743245780467987,
1365
+ "eval_rewards/rejected": -0.027322327718138695,
1366
+ "eval_runtime": 806.7503,
1367
+ "eval_samples_per_second": 2.479,
1368
+ "eval_steps_per_second": 0.31,
1369
+ "step": 955
1370
+ },
1371
+ {
1372
+ "epoch": 1.0,
1373
+ "step": 955,
1374
+ "total_flos": 0.0,
1375
+ "train_loss": 0.6598132096035942,
1376
+ "train_runtime": 45126.4521,
1377
+ "train_samples_per_second": 1.355,
1378
+ "train_steps_per_second": 0.021
1379
+ }
1380
+ ],
1381
+ "logging_steps": 10,
1382
+ "max_steps": 955,
1383
+ "num_train_epochs": 1,
1384
+ "save_steps": 500,
1385
+ "total_flos": 0.0,
1386
+ "trial_name": null,
1387
+ "trial_params": null
1388
+ }
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:39c7c26db1d12916b28e685fe9c90c4503e186a912f65b79dfdb343d089a8828
3
- size 4792
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:55735e718abcb162ead37e5dd69bbbcafb5bd7d79744ef90d96d7c4c7c4f2969
3
+ size 4728