Hugging Face's logo

juspay
/
xyne-rl

Transformers
Safetensors
xyne
grpo
lora-merged
gemma4
Model card Files
xet
 Community
Ashx098 commited on
Commit
80e9f2e
·
verified ·
1 Parent(s): 7e9c3f9

upload adapter-200 adapter files

Browse files
Files changed (1) hide show
  1. v1/adapters/adapter-200/checkpoint-8/trainer_state.json +167 -0
v1/adapters/adapter-200/checkpoint-8/trainer_state.json ADDED
@@ -0,0 +1,167 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 1.0,
6
+ "eval_steps": 500,
7
+ "global_step": 8,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "clip_ratio/high_max": 0.0,
14
+ "clip_ratio/high_mean": 0.0,
15
+ "clip_ratio/low_mean": 0.0,
16
+ "clip_ratio/low_min": 0.0,
17
+ "clip_ratio/region_mean": 0.0,
18
+ "completions/clipped_ratio": 0.0,
19
+ "completions/max_length": 1178.0,
20
+ "completions/max_terminated_length": 1178.0,
21
+ "completions/mean_length": 518.25,
22
+ "completions/mean_terminated_length": 518.25,
23
+ "completions/min_length": 102.0,
24
+ "completions/min_terminated_length": 102.0,
25
+ "entropy": 0.550417423248291,
26
+ "epoch": 0.125,
27
+ "frac_reward_zero_std": 0.1666666716337204,
28
+ "grad_norm": 0.0,
29
+ "kl": 0.0,
30
+ "learning_rate": 1e-06,
31
+ "loss": -0.019361913204193115,
32
+ "num_tokens": 27996.0,
33
+ "reward": 0.5393750071525574,
34
+ "reward_std": 0.3599787652492523,
35
+ "rewards/xyne_reward_fn/mean": 0.5393750071525574,
36
+ "rewards/xyne_reward_fn/std": 0.35997873544692993,
37
+ "step": 1,
38
+ "step_time": 813.1953905383125
39
+ },
40
+ {
41
+ "clip_ratio/high_max": 0.0,
42
+ "clip_ratio/high_mean": 0.0,
43
+ "clip_ratio/low_mean": 0.0,
44
+ "clip_ratio/low_min": 0.0,
45
+ "clip_ratio/region_mean": 0.0,
46
+ "entropy": 0.36864861845970154,
47
+ "epoch": 0.25,
48
+ "grad_norm": 0.0,
49
+ "kl": 0.0,
50
+ "learning_rate": 1e-06,
51
+ "loss": 0.02459419146180153,
52
+ "step": 2,
53
+ "step_time": 0.5488410629332066
54
+ },
55
+ {
56
+ "clip_ratio/high_max": 0.0,
57
+ "clip_ratio/high_mean": 0.0,
58
+ "clip_ratio/low_mean": 0.0,
59
+ "clip_ratio/low_min": 0.0,
60
+ "clip_ratio/region_mean": 0.0,
61
+ "entropy": 0.37100768089294434,
62
+ "epoch": 0.375,
63
+ "grad_norm": 0.0,
64
+ "kl": 0.0,
65
+ "learning_rate": 1e-06,
66
+ "loss": 0.015427134931087494,
67
+ "step": 3,
68
+ "step_time": 0.5428090924397111
69
+ },
70
+ {
71
+ "clip_ratio/high_max": 0.0,
72
+ "clip_ratio/high_mean": 0.0,
73
+ "clip_ratio/low_mean": 0.0,
74
+ "clip_ratio/low_min": 0.0,
75
+ "clip_ratio/region_mean": 0.0,
76
+ "entropy": 0.3826420307159424,
77
+ "epoch": 0.5,
78
+ "grad_norm": 0.0,
79
+ "kl": 0.0,
80
+ "learning_rate": 1e-06,
81
+ "loss": -0.030477195978164673,
82
+ "step": 4,
83
+ "step_time": 0.5439383331686258
84
+ },
85
+ {
86
+ "clip_ratio/high_max": 0.0,
87
+ "clip_ratio/high_mean": 0.0,
88
+ "clip_ratio/low_mean": 0.0,
89
+ "clip_ratio/low_min": 0.0,
90
+ "clip_ratio/region_mean": 0.0,
91
+ "entropy": 0.4469599425792694,
92
+ "epoch": 0.625,
93
+ "grad_norm": 0.0,
94
+ "kl": 0.0,
95
+ "learning_rate": 1e-06,
96
+ "loss": -0.03193400055170059,
97
+ "step": 5,
98
+ "step_time": 0.49569029081612825
99
+ },
100
+ {
101
+ "clip_ratio/high_max": 0.0,
102
+ "clip_ratio/high_mean": 0.0,
103
+ "clip_ratio/low_mean": 0.0,
104
+ "clip_ratio/low_min": 0.0,
105
+ "clip_ratio/region_mean": 0.0,
106
+ "entropy": 0.4180392026901245,
107
+ "epoch": 0.75,
108
+ "grad_norm": 0.0,
109
+ "kl": 0.0,
110
+ "learning_rate": 1e-06,
111
+ "loss": -0.01200093887746334,
112
+ "step": 6,
113
+ "step_time": 0.6537656327709556
114
+ },
115
+ {
116
+ "clip_ratio/high_max": 0.0,
117
+ "clip_ratio/high_mean": 0.0,
118
+ "clip_ratio/low_mean": 0.0,
119
+ "clip_ratio/low_min": 0.0,
120
+ "clip_ratio/region_mean": 0.0,
121
+ "entropy": 0.3941870629787445,
122
+ "epoch": 0.875,
123
+ "grad_norm": 0.0,
124
+ "kl": 0.0,
125
+ "learning_rate": 1e-06,
126
+ "loss": -0.028620531782507896,
127
+ "step": 7,
128
+ "step_time": 0.554437174461782
129
+ },
130
+ {
131
+ "clip_ratio/high_max": 0.0,
132
+ "clip_ratio/high_mean": 0.0,
133
+ "clip_ratio/low_mean": 0.0,
134
+ "clip_ratio/low_min": 0.0,
135
+ "clip_ratio/region_mean": 0.0,
136
+ "entropy": 0.36311957240104675,
137
+ "epoch": 1.0,
138
+ "grad_norm": 0.0,
139
+ "kl": 0.0,
140
+ "learning_rate": 1e-06,
141
+ "loss": -0.018424350768327713,
142
+ "step": 8,
143
+ "step_time": 0.5464225821197033
144
+ }
145
+ ],
146
+ "logging_steps": 1,
147
+ "max_steps": 8,
148
+ "num_input_tokens_seen": 27996,
149
+ "num_train_epochs": 1,
150
+ "save_steps": 8,
151
+ "stateful_callbacks": {
152
+ "TrainerControl": {
153
+ "args": {
154
+ "should_epoch_stop": false,
155
+ "should_evaluate": false,
156
+ "should_log": false,
157
+ "should_save": true,
158
+ "should_training_stop": true
159
+ },
160
+ "attributes": {}
161
+ }
162
+ },
163
+ "total_flos": 0.0,
164
+ "train_batch_size": 1,
165
+ "trial_name": null,
166
+ "trial_params": null
167
+ }