End of training
Browse files- all_results.json +5 -5
- model.safetensors +1 -1
- train_results.json +5 -5
- trainer_state.json +131 -131
all_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"total_flos": 0.0,
|
| 3 |
-
"train_loss": 0.
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples":
|
| 6 |
-
"train_samples_per_second": 0.
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"total_flos": 0.0,
|
| 3 |
+
"train_loss": 0.00457829560036771,
|
| 4 |
+
"train_runtime": 805.5031,
|
| 5 |
+
"train_samples": 161,
|
| 6 |
+
"train_samples_per_second": 0.397,
|
| 7 |
+
"train_steps_per_second": 0.025
|
| 8 |
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1976163472
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:25f4be18da37ed009b3a825b1861d6c27a947952f19a732fb0d1720782c18587
|
| 3 |
size 1976163472
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"total_flos": 0.0,
|
| 3 |
-
"train_loss": 0.
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples":
|
| 6 |
-
"train_samples_per_second": 0.
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"total_flos": 0.0,
|
| 3 |
+
"train_loss": 0.00457829560036771,
|
| 4 |
+
"train_runtime": 805.5031,
|
| 5 |
+
"train_samples": 161,
|
| 6 |
+
"train_samples_per_second": 0.397,
|
| 7 |
+
"train_steps_per_second": 0.025
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -2,7 +2,7 @@
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
-
"epoch": 0.
|
| 6 |
"eval_steps": 500,
|
| 7 |
"global_step": 20,
|
| 8 |
"is_hyper_param_search": false,
|
|
@@ -10,203 +10,203 @@
|
|
| 10 |
"is_world_process_zero": true,
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
-
"completion_length":
|
| 14 |
-
"epoch": 0.
|
| 15 |
-
"grad_norm":
|
| 16 |
"kl": 0.0,
|
| 17 |
"learning_rate": 4.965903258506806e-07,
|
| 18 |
-
"loss": 0.0,
|
| 19 |
-
"reward":
|
| 20 |
-
"reward_std": 0.
|
| 21 |
-
"rewards/concensus_correctness_reward_func": 0.
|
| 22 |
-
"rewards/consensus_reward_func": 0.
|
| 23 |
"rewards/cumulative_reward_2": 0.0,
|
| 24 |
-
"rewards/final_correctness_reward_func": 0.
|
| 25 |
-
"rewards/question_recreation_reward_func": 0.
|
| 26 |
"rewards/soft_format_reward_func": 0.0,
|
| 27 |
-
"rewards/strict_format_reward_func": 0.
|
| 28 |
-
"rewards/xmlcount_reward_func": 0.
|
| 29 |
"step": 2
|
| 30 |
},
|
| 31 |
{
|
| 32 |
-
"completion_length":
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"kl": 0.
|
| 36 |
"learning_rate": 4.698684378016222e-07,
|
| 37 |
"loss": 0.0,
|
| 38 |
-
"reward": 3.
|
| 39 |
-
"reward_std": 1.
|
| 40 |
-
"rewards/concensus_correctness_reward_func": 0.
|
| 41 |
-
"rewards/consensus_reward_func": 0.
|
| 42 |
"rewards/cumulative_reward_2": 0.0,
|
| 43 |
-
"rewards/final_correctness_reward_func": 0.
|
| 44 |
-
"rewards/question_recreation_reward_func": 0.
|
| 45 |
"rewards/soft_format_reward_func": 0.0,
|
| 46 |
-
"rewards/strict_format_reward_func": 0.
|
| 47 |
-
"rewards/xmlcount_reward_func": 1.
|
| 48 |
"step": 4
|
| 49 |
},
|
| 50 |
{
|
| 51 |
-
"completion_length":
|
| 52 |
-
"epoch": 0.
|
| 53 |
-
"grad_norm":
|
| 54 |
-
"kl": 0.
|
| 55 |
"learning_rate": 4.193203929064353e-07,
|
| 56 |
-
"loss": 0.
|
| 57 |
-
"reward": 3.
|
| 58 |
-
"reward_std": 1.
|
| 59 |
-
"rewards/concensus_correctness_reward_func":
|
| 60 |
-
"rewards/consensus_reward_func": 0.
|
| 61 |
"rewards/cumulative_reward_2": 0.0,
|
| 62 |
-
"rewards/final_correctness_reward_func": 0.
|
| 63 |
-
"rewards/question_recreation_reward_func": 0.
|
| 64 |
"rewards/soft_format_reward_func": 0.0,
|
| 65 |
-
"rewards/strict_format_reward_func": 0.
|
| 66 |
-
"rewards/xmlcount_reward_func": 1.
|
| 67 |
"step": 6
|
| 68 |
},
|
| 69 |
{
|
| 70 |
-
"completion_length":
|
| 71 |
-
"epoch": 0.
|
| 72 |
-
"grad_norm":
|
| 73 |
-
"kl": 0.
|
| 74 |
"learning_rate": 3.5042385616324236e-07,
|
| 75 |
"loss": 0.0002,
|
| 76 |
-
"reward":
|
| 77 |
-
"reward_std":
|
| 78 |
-
"rewards/concensus_correctness_reward_func":
|
| 79 |
-
"rewards/consensus_reward_func":
|
| 80 |
"rewards/cumulative_reward_2": 0.0,
|
| 81 |
-
"rewards/final_correctness_reward_func": 0.
|
| 82 |
-
"rewards/question_recreation_reward_func": 0.
|
| 83 |
"rewards/soft_format_reward_func": 0.0,
|
| 84 |
-
"rewards/strict_format_reward_func": 0.
|
| 85 |
-
"rewards/xmlcount_reward_func": 1.
|
| 86 |
"step": 8
|
| 87 |
},
|
| 88 |
{
|
| 89 |
-
"completion_length":
|
| 90 |
-
"epoch": 0.
|
| 91 |
-
"grad_norm":
|
| 92 |
-
"kl": 1.
|
| 93 |
"learning_rate": 2.706448363680831e-07,
|
| 94 |
-
"loss": 0.
|
| 95 |
-
"reward":
|
| 96 |
-
"reward_std":
|
| 97 |
-
"rewards/concensus_correctness_reward_func": 1.
|
| 98 |
-
"rewards/consensus_reward_func": 0.
|
| 99 |
"rewards/cumulative_reward_2": 0.0,
|
| 100 |
-
"rewards/final_correctness_reward_func": 0.
|
| 101 |
-
"rewards/question_recreation_reward_func": 0.
|
| 102 |
"rewards/soft_format_reward_func": 0.0,
|
| 103 |
-
"rewards/strict_format_reward_func": 0.
|
| 104 |
-
"rewards/xmlcount_reward_func": 1.
|
| 105 |
"step": 10
|
| 106 |
},
|
| 107 |
{
|
| 108 |
-
"completion_length":
|
| 109 |
-
"epoch": 0.
|
| 110 |
-
"grad_norm":
|
| 111 |
-
"kl":
|
| 112 |
"learning_rate": 1.886286282148002e-07,
|
| 113 |
-
"loss": 0.
|
| 114 |
-
"reward": 3.
|
| 115 |
-
"reward_std":
|
| 116 |
-
"rewards/concensus_correctness_reward_func": 0.
|
| 117 |
-
"rewards/consensus_reward_func": 0.
|
| 118 |
"rewards/cumulative_reward_2": 0.0,
|
| 119 |
-
"rewards/final_correctness_reward_func": 0.
|
| 120 |
-
"rewards/question_recreation_reward_func": 0.
|
| 121 |
"rewards/soft_format_reward_func": 0.0,
|
| 122 |
-
"rewards/strict_format_reward_func": 0.
|
| 123 |
-
"rewards/xmlcount_reward_func": 1.
|
| 124 |
"step": 12
|
| 125 |
},
|
| 126 |
{
|
| 127 |
-
"completion_length":
|
| 128 |
-
"epoch": 0.
|
| 129 |
-
"grad_norm":
|
| 130 |
-
"kl":
|
| 131 |
"learning_rate": 1.1326296046939333e-07,
|
| 132 |
-
"loss": 0.
|
| 133 |
-
"reward":
|
| 134 |
-
"reward_std": 2.
|
| 135 |
-
"rewards/concensus_correctness_reward_func":
|
| 136 |
"rewards/consensus_reward_func": 0.3125,
|
| 137 |
"rewards/cumulative_reward_2": 0.0,
|
| 138 |
-
"rewards/final_correctness_reward_func": 0.
|
| 139 |
-
"rewards/question_recreation_reward_func": 0.
|
| 140 |
"rewards/soft_format_reward_func": 0.0,
|
| 141 |
-
"rewards/strict_format_reward_func": 0.
|
| 142 |
-
"rewards/xmlcount_reward_func":
|
| 143 |
"step": 14
|
| 144 |
},
|
| 145 |
{
|
| 146 |
-
"completion_length":
|
| 147 |
-
"epoch": 0.
|
| 148 |
-
"grad_norm":
|
| 149 |
-
"kl": 0.
|
| 150 |
"learning_rate": 5.271487265090163e-08,
|
| 151 |
-
"loss": 0.
|
| 152 |
-
"reward": 5.
|
| 153 |
-
"reward_std":
|
| 154 |
-
"rewards/concensus_correctness_reward_func": 2.
|
| 155 |
-
"rewards/consensus_reward_func":
|
| 156 |
"rewards/cumulative_reward_2": 0.0,
|
| 157 |
-
"rewards/final_correctness_reward_func": 0.
|
| 158 |
-
"rewards/question_recreation_reward_func": 0.
|
| 159 |
"rewards/soft_format_reward_func": 0.0,
|
| 160 |
-
"rewards/strict_format_reward_func": 0.
|
| 161 |
-
"rewards/xmlcount_reward_func": 1.
|
| 162 |
"step": 16
|
| 163 |
},
|
| 164 |
{
|
| 165 |
-
"completion_length":
|
| 166 |
-
"epoch": 0.
|
| 167 |
-
"grad_norm":
|
| 168 |
-
"kl":
|
| 169 |
"learning_rate": 1.3545689574841341e-08,
|
| 170 |
-
"loss": 0.
|
| 171 |
-
"reward":
|
| 172 |
-
"reward_std":
|
| 173 |
-
"rewards/concensus_correctness_reward_func":
|
| 174 |
-
"rewards/consensus_reward_func": 0.
|
| 175 |
"rewards/cumulative_reward_2": 0.0,
|
| 176 |
-
"rewards/final_correctness_reward_func": 0.
|
| 177 |
-
"rewards/question_recreation_reward_func": 0.
|
| 178 |
"rewards/soft_format_reward_func": 0.0,
|
| 179 |
-
"rewards/strict_format_reward_func": 0.
|
| 180 |
-
"rewards/xmlcount_reward_func":
|
| 181 |
"step": 18
|
| 182 |
},
|
| 183 |
{
|
| 184 |
-
"completion_length":
|
| 185 |
-
"epoch": 0.
|
| 186 |
-
"grad_norm":
|
| 187 |
-
"kl":
|
| 188 |
"learning_rate": 0.0,
|
| 189 |
-
"loss": 0.
|
| 190 |
-
"reward": 4.
|
| 191 |
-
"reward_std":
|
| 192 |
-
"rewards/concensus_correctness_reward_func": 1.
|
| 193 |
-
"rewards/consensus_reward_func": 0.
|
| 194 |
"rewards/cumulative_reward_2": 0.0,
|
| 195 |
"rewards/final_correctness_reward_func": 0.5625,
|
| 196 |
-
"rewards/question_recreation_reward_func": 0.
|
| 197 |
"rewards/soft_format_reward_func": 0.0,
|
| 198 |
-
"rewards/strict_format_reward_func": 0.
|
| 199 |
-
"rewards/xmlcount_reward_func":
|
| 200 |
"step": 20
|
| 201 |
},
|
| 202 |
{
|
| 203 |
-
"epoch": 0.
|
| 204 |
"step": 20,
|
| 205 |
"total_flos": 0.0,
|
| 206 |
-
"train_loss": 0.
|
| 207 |
-
"train_runtime":
|
| 208 |
-
"train_samples_per_second": 0.
|
| 209 |
-
"train_steps_per_second": 0.
|
| 210 |
}
|
| 211 |
],
|
| 212 |
"logging_steps": 2,
|
|
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.9937888198757764,
|
| 6 |
"eval_steps": 500,
|
| 7 |
"global_step": 20,
|
| 8 |
"is_hyper_param_search": false,
|
|
|
|
| 10 |
"is_world_process_zero": true,
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
+
"completion_length": 311.90625,
|
| 14 |
+
"epoch": 0.09937888198757763,
|
| 15 |
+
"grad_norm": 16.116872787475586,
|
| 16 |
"kl": 0.0,
|
| 17 |
"learning_rate": 4.965903258506806e-07,
|
| 18 |
+
"loss": -0.0,
|
| 19 |
+
"reward": 2.0880651086336,
|
| 20 |
+
"reward_std": 0.6684938869730104,
|
| 21 |
+
"rewards/concensus_correctness_reward_func": 0.3713750019669533,
|
| 22 |
+
"rewards/consensus_reward_func": 0.3125,
|
| 23 |
"rewards/cumulative_reward_2": 0.0,
|
| 24 |
+
"rewards/final_correctness_reward_func": 0.3125,
|
| 25 |
+
"rewards/question_recreation_reward_func": 0.39662764465902,
|
| 26 |
"rewards/soft_format_reward_func": 0.0,
|
| 27 |
+
"rewards/strict_format_reward_func": 0.15625,
|
| 28 |
+
"rewards/xmlcount_reward_func": 0.5388124994933605,
|
| 29 |
"step": 2
|
| 30 |
},
|
| 31 |
{
|
| 32 |
+
"completion_length": 187.03125,
|
| 33 |
+
"epoch": 0.19875776397515527,
|
| 34 |
+
"grad_norm": 20.159204483032227,
|
| 35 |
+
"kl": 0.03585821596789174,
|
| 36 |
"learning_rate": 4.698684378016222e-07,
|
| 37 |
"loss": 0.0,
|
| 38 |
+
"reward": 3.9214092064648867,
|
| 39 |
+
"reward_std": 1.652882975889952,
|
| 40 |
+
"rewards/concensus_correctness_reward_func": 0.9944375082850456,
|
| 41 |
+
"rewards/consensus_reward_func": 0.5625,
|
| 42 |
"rewards/cumulative_reward_2": 0.0,
|
| 43 |
+
"rewards/final_correctness_reward_func": 0.25,
|
| 44 |
+
"rewards/question_recreation_reward_func": 0.5815029521472752,
|
| 45 |
"rewards/soft_format_reward_func": 0.0,
|
| 46 |
+
"rewards/strict_format_reward_func": 0.359375,
|
| 47 |
+
"rewards/xmlcount_reward_func": 1.173593744635582,
|
| 48 |
"step": 4
|
| 49 |
},
|
| 50 |
{
|
| 51 |
+
"completion_length": 179.375,
|
| 52 |
+
"epoch": 0.2981366459627329,
|
| 53 |
+
"grad_norm": 26.64881134033203,
|
| 54 |
+
"kl": 0.47550190216861665,
|
| 55 |
"learning_rate": 4.193203929064353e-07,
|
| 56 |
+
"loss": 0.0005,
|
| 57 |
+
"reward": 3.2665678560733795,
|
| 58 |
+
"reward_std": 1.333875675802119,
|
| 59 |
+
"rewards/concensus_correctness_reward_func": 0.4598749987781048,
|
| 60 |
+
"rewards/consensus_reward_func": 0.4375,
|
| 61 |
"rewards/cumulative_reward_2": 0.0,
|
| 62 |
+
"rewards/final_correctness_reward_func": 0.375,
|
| 63 |
+
"rewards/question_recreation_reward_func": 0.48253656178712845,
|
| 64 |
"rewards/soft_format_reward_func": 0.0,
|
| 65 |
+
"rewards/strict_format_reward_func": 0.390625,
|
| 66 |
+
"rewards/xmlcount_reward_func": 1.121031254529953,
|
| 67 |
"step": 6
|
| 68 |
},
|
| 69 |
{
|
| 70 |
+
"completion_length": 206.5625,
|
| 71 |
+
"epoch": 0.39751552795031053,
|
| 72 |
+
"grad_norm": 28.57688331604004,
|
| 73 |
+
"kl": 0.20366718503646553,
|
| 74 |
"learning_rate": 3.5042385616324236e-07,
|
| 75 |
"loss": 0.0002,
|
| 76 |
+
"reward": 5.219966318458319,
|
| 77 |
+
"reward_std": 2.007999164707144,
|
| 78 |
+
"rewards/concensus_correctness_reward_func": 1.7436874955892563,
|
| 79 |
+
"rewards/consensus_reward_func": 1.0625,
|
| 80 |
"rewards/cumulative_reward_2": 0.0,
|
| 81 |
+
"rewards/final_correctness_reward_func": 0.3125,
|
| 82 |
+
"rewards/question_recreation_reward_func": 0.6211538150673732,
|
| 83 |
"rewards/soft_format_reward_func": 0.0,
|
| 84 |
+
"rewards/strict_format_reward_func": 0.34375,
|
| 85 |
+
"rewards/xmlcount_reward_func": 1.1363750025629997,
|
| 86 |
"step": 8
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"completion_length": 169.3125,
|
| 90 |
+
"epoch": 0.4968944099378882,
|
| 91 |
+
"grad_norm": 44.238037109375,
|
| 92 |
+
"kl": 1.7451796566601843,
|
| 93 |
"learning_rate": 2.706448363680831e-07,
|
| 94 |
+
"loss": 0.0017,
|
| 95 |
+
"reward": 4.474540323019028,
|
| 96 |
+
"reward_std": 1.2630424201197457,
|
| 97 |
+
"rewards/concensus_correctness_reward_func": 1.2620625011622906,
|
| 98 |
+
"rewards/consensus_reward_func": 0.625,
|
| 99 |
"rewards/cumulative_reward_2": 0.0,
|
| 100 |
+
"rewards/final_correctness_reward_func": 0.5625,
|
| 101 |
+
"rewards/question_recreation_reward_func": 0.6150090787559748,
|
| 102 |
"rewards/soft_format_reward_func": 0.0,
|
| 103 |
+
"rewards/strict_format_reward_func": 0.296875,
|
| 104 |
+
"rewards/xmlcount_reward_func": 1.1130937486886978,
|
| 105 |
"step": 10
|
| 106 |
},
|
| 107 |
{
|
| 108 |
+
"completion_length": 206.59375,
|
| 109 |
+
"epoch": 0.5962732919254659,
|
| 110 |
+
"grad_norm": 45.16315841674805,
|
| 111 |
+
"kl": 2.844578029587865,
|
| 112 |
"learning_rate": 1.886286282148002e-07,
|
| 113 |
+
"loss": 0.0028,
|
| 114 |
+
"reward": 3.3002478405833244,
|
| 115 |
+
"reward_std": 1.4689021888189018,
|
| 116 |
+
"rewards/concensus_correctness_reward_func": 0.6563750021159649,
|
| 117 |
+
"rewards/consensus_reward_func": 0.5,
|
| 118 |
"rewards/cumulative_reward_2": 0.0,
|
| 119 |
+
"rewards/final_correctness_reward_func": 0.1875,
|
| 120 |
+
"rewards/question_recreation_reward_func": 0.5149665717035532,
|
| 121 |
"rewards/soft_format_reward_func": 0.0,
|
| 122 |
+
"rewards/strict_format_reward_func": 0.328125,
|
| 123 |
+
"rewards/xmlcount_reward_func": 1.11328125,
|
| 124 |
"step": 12
|
| 125 |
},
|
| 126 |
{
|
| 127 |
+
"completion_length": 199.78125,
|
| 128 |
+
"epoch": 0.6956521739130435,
|
| 129 |
+
"grad_norm": 703.441650390625,
|
| 130 |
+
"kl": 30.634997794870287,
|
| 131 |
"learning_rate": 1.1326296046939333e-07,
|
| 132 |
+
"loss": 0.0306,
|
| 133 |
+
"reward": 4.00101051479578,
|
| 134 |
+
"reward_std": 2.141244651051238,
|
| 135 |
+
"rewards/concensus_correctness_reward_func": 1.3625624813139439,
|
| 136 |
"rewards/consensus_reward_func": 0.3125,
|
| 137 |
"rewards/cumulative_reward_2": 0.0,
|
| 138 |
+
"rewards/final_correctness_reward_func": 0.3125,
|
| 139 |
+
"rewards/question_recreation_reward_func": 0.5683229928836226,
|
| 140 |
"rewards/soft_format_reward_func": 0.0,
|
| 141 |
+
"rewards/strict_format_reward_func": 0.296875,
|
| 142 |
+
"rewards/xmlcount_reward_func": 1.1482499986886978,
|
| 143 |
"step": 14
|
| 144 |
},
|
| 145 |
{
|
| 146 |
+
"completion_length": 189.5625,
|
| 147 |
+
"epoch": 0.7950310559006211,
|
| 148 |
+
"grad_norm": 22.315275192260742,
|
| 149 |
+
"kl": 0.6838863966986537,
|
| 150 |
"learning_rate": 5.271487265090163e-08,
|
| 151 |
+
"loss": 0.0007,
|
| 152 |
+
"reward": 5.960442878305912,
|
| 153 |
+
"reward_std": 1.6150732710957527,
|
| 154 |
+
"rewards/concensus_correctness_reward_func": 2.5391249880194664,
|
| 155 |
+
"rewards/consensus_reward_func": 0.8125,
|
| 156 |
"rewards/cumulative_reward_2": 0.0,
|
| 157 |
+
"rewards/final_correctness_reward_func": 0.625,
|
| 158 |
+
"rewards/question_recreation_reward_func": 0.5619429536163807,
|
| 159 |
"rewards/soft_format_reward_func": 0.0,
|
| 160 |
+
"rewards/strict_format_reward_func": 0.328125,
|
| 161 |
+
"rewards/xmlcount_reward_func": 1.09375,
|
| 162 |
"step": 16
|
| 163 |
},
|
| 164 |
{
|
| 165 |
+
"completion_length": 197.75,
|
| 166 |
+
"epoch": 0.8944099378881988,
|
| 167 |
+
"grad_norm": 52.568763732910156,
|
| 168 |
+
"kl": 2.449893494369462,
|
| 169 |
"learning_rate": 1.3545689574841341e-08,
|
| 170 |
+
"loss": 0.0024,
|
| 171 |
+
"reward": 4.1809456795454025,
|
| 172 |
+
"reward_std": 2.145681773661636,
|
| 173 |
+
"rewards/concensus_correctness_reward_func": 1.2196249850094318,
|
| 174 |
+
"rewards/consensus_reward_func": 0.5625,
|
| 175 |
"rewards/cumulative_reward_2": 0.0,
|
| 176 |
+
"rewards/final_correctness_reward_func": 0.3125,
|
| 177 |
+
"rewards/question_recreation_reward_func": 0.582414451870136,
|
| 178 |
"rewards/soft_format_reward_func": 0.0,
|
| 179 |
+
"rewards/strict_format_reward_func": 0.359375,
|
| 180 |
+
"rewards/xmlcount_reward_func": 1.14453125,
|
| 181 |
"step": 18
|
| 182 |
},
|
| 183 |
{
|
| 184 |
+
"completion_length": 247.875,
|
| 185 |
+
"epoch": 0.9937888198757764,
|
| 186 |
+
"grad_norm": 106.65875244140625,
|
| 187 |
+
"kl": 6.712249951437116,
|
| 188 |
"learning_rate": 0.0,
|
| 189 |
+
"loss": 0.0067,
|
| 190 |
+
"reward": 4.293041463941336,
|
| 191 |
+
"reward_std": 1.9913227264769375,
|
| 192 |
+
"rewards/concensus_correctness_reward_func": 1.4836874902248383,
|
| 193 |
+
"rewards/consensus_reward_func": 0.4375,
|
| 194 |
"rewards/cumulative_reward_2": 0.0,
|
| 195 |
"rewards/final_correctness_reward_func": 0.5625,
|
| 196 |
+
"rewards/question_recreation_reward_func": 0.5593539979308844,
|
| 197 |
"rewards/soft_format_reward_func": 0.0,
|
| 198 |
+
"rewards/strict_format_reward_func": 0.265625,
|
| 199 |
+
"rewards/xmlcount_reward_func": 0.984375,
|
| 200 |
"step": 20
|
| 201 |
},
|
| 202 |
{
|
| 203 |
+
"epoch": 0.9937888198757764,
|
| 204 |
"step": 20,
|
| 205 |
"total_flos": 0.0,
|
| 206 |
+
"train_loss": 0.00457829560036771,
|
| 207 |
+
"train_runtime": 805.5031,
|
| 208 |
+
"train_samples_per_second": 0.397,
|
| 209 |
+
"train_steps_per_second": 0.025
|
| 210 |
}
|
| 211 |
],
|
| 212 |
"logging_steps": 2,
|