File size: 8,909 Bytes
5ef9bc7 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 |
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.9984,
"eval_steps": 100,
"global_step": 156,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 3.125e-07,
"logits/chosen": -2.7403299808502197,
"logits/rejected": -2.7255897521972656,
"logps/chosen": -195.21282958984375,
"logps/rejected": -184.09413146972656,
"loss": 0.6931,
"rewards/accuracies": 0.0,
"rewards/chosen": 0.0,
"rewards/margins": 0.0,
"rewards/rejected": 0.0,
"step": 1
},
{
"epoch": 0.06,
"learning_rate": 3.125e-06,
"logits/chosen": -2.6841495037078857,
"logits/rejected": -2.682373523712158,
"logps/chosen": -163.6064453125,
"logps/rejected": -165.0059356689453,
"loss": 0.6929,
"rewards/accuracies": 0.4756944477558136,
"rewards/chosen": -0.008276204578578472,
"rewards/margins": 0.001318673719651997,
"rewards/rejected": -0.009594877250492573,
"step": 10
},
{
"epoch": 0.13,
"learning_rate": 4.989935734988098e-06,
"logits/chosen": -2.692333698272705,
"logits/rejected": -2.665407180786133,
"logps/chosen": -176.43678283691406,
"logps/rejected": -172.32717895507812,
"loss": 0.6935,
"rewards/accuracies": 0.4906249940395355,
"rewards/chosen": -0.06923351436853409,
"rewards/margins": 0.003520409809425473,
"rewards/rejected": -0.0727539211511612,
"step": 20
},
{
"epoch": 0.19,
"learning_rate": 4.8776412907378845e-06,
"logits/chosen": -2.5914735794067383,
"logits/rejected": -2.585202932357788,
"logps/chosen": -168.06890869140625,
"logps/rejected": -169.85023498535156,
"loss": 0.6917,
"rewards/accuracies": 0.550000011920929,
"rewards/chosen": -0.034951962530612946,
"rewards/margins": 0.007884470745921135,
"rewards/rejected": -0.04283643513917923,
"step": 30
},
{
"epoch": 0.26,
"learning_rate": 4.646121984004666e-06,
"logits/chosen": -2.671989917755127,
"logits/rejected": -2.6600332260131836,
"logps/chosen": -178.48281860351562,
"logps/rejected": -179.00379943847656,
"loss": 0.6901,
"rewards/accuracies": 0.53125,
"rewards/chosen": -0.01323291938751936,
"rewards/margins": 0.006769413594156504,
"rewards/rejected": -0.020002331584692,
"step": 40
},
{
"epoch": 0.32,
"learning_rate": 4.3069871595684795e-06,
"logits/chosen": -2.6603026390075684,
"logits/rejected": -2.6338632106781006,
"logps/chosen": -188.98532104492188,
"logps/rejected": -187.17684936523438,
"loss": 0.6888,
"rewards/accuracies": 0.5406249761581421,
"rewards/chosen": -0.09863803535699844,
"rewards/margins": 0.015061549842357635,
"rewards/rejected": -0.11369959264993668,
"step": 50
},
{
"epoch": 0.38,
"learning_rate": 3.8772424536302565e-06,
"logits/chosen": -2.6301541328430176,
"logits/rejected": -2.6398401260375977,
"logps/chosen": -184.1541290283203,
"logps/rejected": -188.9746856689453,
"loss": 0.6898,
"rewards/accuracies": 0.543749988079071,
"rewards/chosen": -0.16303817927837372,
"rewards/margins": 0.013844095170497894,
"rewards/rejected": -0.1768822818994522,
"step": 60
},
{
"epoch": 0.45,
"learning_rate": 3.3784370602033572e-06,
"logits/chosen": -2.587550401687622,
"logits/rejected": -2.550933361053467,
"logps/chosen": -184.53257751464844,
"logps/rejected": -183.46517944335938,
"loss": 0.6902,
"rewards/accuracies": 0.4937500059604645,
"rewards/chosen": -0.008703355677425861,
"rewards/margins": 0.0026464841794222593,
"rewards/rejected": -0.011349838227033615,
"step": 70
},
{
"epoch": 0.51,
"learning_rate": 2.835583164544139e-06,
"logits/chosen": -2.56795072555542,
"logits/rejected": -2.5637052059173584,
"logps/chosen": -176.33071899414062,
"logps/rejected": -184.88938903808594,
"loss": 0.6906,
"rewards/accuracies": 0.578125,
"rewards/chosen": 0.0037323920987546444,
"rewards/margins": 0.009280242025852203,
"rewards/rejected": -0.005547848995774984,
"step": 80
},
{
"epoch": 0.58,
"learning_rate": 2.2759017277414165e-06,
"logits/chosen": -2.6137020587921143,
"logits/rejected": -2.5704140663146973,
"logps/chosen": -187.82107543945312,
"logps/rejected": -184.28587341308594,
"loss": 0.6883,
"rewards/accuracies": 0.5375000238418579,
"rewards/chosen": -0.08189593255519867,
"rewards/margins": 0.0156090734526515,
"rewards/rejected": -0.09750500321388245,
"step": 90
},
{
"epoch": 0.64,
"learning_rate": 1.7274575140626318e-06,
"logits/chosen": -2.6152119636535645,
"logits/rejected": -2.640665292739868,
"logps/chosen": -176.8973388671875,
"logps/rejected": -184.32557678222656,
"loss": 0.689,
"rewards/accuracies": 0.53125,
"rewards/chosen": -0.08713572472333908,
"rewards/margins": 0.014782750979065895,
"rewards/rejected": -0.10191845893859863,
"step": 100
},
{
"epoch": 0.64,
"eval_logits/chosen": -2.556511878967285,
"eval_logits/rejected": -2.465257406234741,
"eval_logps/chosen": -305.5672607421875,
"eval_logps/rejected": -298.1529541015625,
"eval_loss": 0.6753088235855103,
"eval_rewards/accuracies": 0.6359999775886536,
"eval_rewards/chosen": -0.11736557632684708,
"eval_rewards/margins": 0.04772435873746872,
"eval_rewards/rejected": -0.1650899201631546,
"eval_runtime": 384.6543,
"eval_samples_per_second": 5.199,
"eval_steps_per_second": 0.65,
"step": 100
},
{
"epoch": 0.7,
"learning_rate": 1.217751806485235e-06,
"logits/chosen": -2.62018084526062,
"logits/rejected": -2.610757350921631,
"logps/chosen": -171.35316467285156,
"logps/rejected": -176.16299438476562,
"loss": 0.6845,
"rewards/accuracies": 0.534375011920929,
"rewards/chosen": -0.05986649543046951,
"rewards/margins": 0.020178433507680893,
"rewards/rejected": -0.0800449326634407,
"step": 110
},
{
"epoch": 0.77,
"learning_rate": 7.723433775328385e-07,
"logits/chosen": -2.548755168914795,
"logits/rejected": -2.529273271560669,
"logps/chosen": -173.8794403076172,
"logps/rejected": -172.54251098632812,
"loss": 0.6854,
"rewards/accuracies": 0.559374988079071,
"rewards/chosen": -0.06846034526824951,
"rewards/margins": 0.019297173246741295,
"rewards/rejected": -0.08775752037763596,
"step": 120
},
{
"epoch": 0.83,
"learning_rate": 4.1356686569674344e-07,
"logits/chosen": -2.6056151390075684,
"logits/rejected": -2.5934414863586426,
"logps/chosen": -181.0892791748047,
"logps/rejected": -185.13357543945312,
"loss": 0.686,
"rewards/accuracies": 0.5562499761581421,
"rewards/chosen": -0.055159904062747955,
"rewards/margins": 0.01887853816151619,
"rewards/rejected": -0.07403843104839325,
"step": 130
},
{
"epoch": 0.9,
"learning_rate": 1.59412823400657e-07,
"logits/chosen": -2.592637538909912,
"logits/rejected": -2.583139657974243,
"logps/chosen": -176.57772827148438,
"logps/rejected": -182.9418182373047,
"loss": 0.6847,
"rewards/accuracies": 0.581250011920929,
"rewards/chosen": -0.04002287983894348,
"rewards/margins": 0.02553923986852169,
"rewards/rejected": -0.06556212902069092,
"step": 140
},
{
"epoch": 0.96,
"learning_rate": 2.262559558016325e-08,
"logits/chosen": -2.5700881481170654,
"logits/rejected": -2.561204433441162,
"logps/chosen": -178.7554473876953,
"logps/rejected": -186.81182861328125,
"loss": 0.6849,
"rewards/accuracies": 0.503125011920929,
"rewards/chosen": -0.05323126167058945,
"rewards/margins": 0.007863727398216724,
"rewards/rejected": -0.0610949881374836,
"step": 150
},
{
"epoch": 1.0,
"step": 156,
"total_flos": 0.0,
"train_loss": 0.6885996239307599,
"train_runtime": 7326.9096,
"train_samples_per_second": 2.73,
"train_steps_per_second": 0.021
}
],
"logging_steps": 10,
"max_steps": 156,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 100,
"total_flos": 0.0,
"train_batch_size": 2,
"trial_name": null,
"trial_params": null
}
|