File size: 7,033 Bytes
466c48a |
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 |
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.6019563581640331,
"eval_steps": 30,
"global_step": 50,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.012039127163280662,
"grad_norm": 11.953282356262207,
"learning_rate": 5.555555555555555e-08,
"logits/chosen": -0.48816660046577454,
"logits/rejected": -0.42142170667648315,
"logps/chosen": -117.26611328125,
"logps/rejected": -125.41987609863281,
"loss": 0.6931,
"rewards/accuracies": 0.0,
"rewards/chosen": 0.0,
"rewards/margins": 0.0,
"rewards/rejected": 0.0,
"step": 1
},
{
"epoch": 0.06019563581640331,
"grad_norm": 16.68506622314453,
"learning_rate": 2.7777777777777776e-07,
"logits/chosen": -0.46595269441604614,
"logits/rejected": -0.356529176235199,
"logps/chosen": -190.95057678222656,
"logps/rejected": -211.25076293945312,
"loss": 0.6926,
"rewards/accuracies": 0.453125,
"rewards/chosen": 0.0007588082225993276,
"rewards/margins": 0.0022044419310986996,
"rewards/rejected": -0.0014456338249146938,
"step": 5
},
{
"epoch": 0.12039127163280662,
"grad_norm": 13.722668647766113,
"learning_rate": 4.999499509357132e-07,
"logits/chosen": -0.4793759286403656,
"logits/rejected": -0.37052756547927856,
"logps/chosen": -155.6678009033203,
"logps/rejected": -199.44947814941406,
"loss": 0.6889,
"rewards/accuracies": 0.8125,
"rewards/chosen": 0.005186144262552261,
"rewards/margins": 0.009383995085954666,
"rewards/rejected": -0.004197851754724979,
"step": 10
},
{
"epoch": 0.18058690744920994,
"grad_norm": 12.493337631225586,
"learning_rate": 4.982003369106287e-07,
"logits/chosen": -0.49185729026794434,
"logits/rejected": -0.37670475244522095,
"logps/chosen": -76.30625915527344,
"logps/rejected": -177.40817260742188,
"loss": 0.6691,
"rewards/accuracies": 1.0,
"rewards/chosen": 0.028129320591688156,
"rewards/margins": 0.048517487943172455,
"rewards/rejected": -0.020388163626194,
"step": 15
},
{
"epoch": 0.24078254326561324,
"grad_norm": 11.79749870300293,
"learning_rate": 4.939682729058838e-07,
"logits/chosen": -0.45126277208328247,
"logits/rejected": -0.3729521930217743,
"logps/chosen": -166.20733642578125,
"logps/rejected": -207.5035400390625,
"loss": 0.6249,
"rewards/accuracies": 1.0,
"rewards/chosen": 0.08626963198184967,
"rewards/margins": 0.14315639436244965,
"rewards/rejected": -0.05688678100705147,
"step": 20
},
{
"epoch": 0.3009781790820166,
"grad_norm": 10.574383735656738,
"learning_rate": 4.872960871766826e-07,
"logits/chosen": -0.4710594713687897,
"logits/rejected": -0.3694532513618469,
"logps/chosen": -86.93299865722656,
"logps/rejected": -186.15248107910156,
"loss": 0.587,
"rewards/accuracies": 1.0,
"rewards/chosen": 0.12460210174322128,
"rewards/margins": 0.23137669265270233,
"rewards/rejected": -0.10677458345890045,
"step": 25
},
{
"epoch": 0.3611738148984199,
"grad_norm": 11.093204498291016,
"learning_rate": 4.782505135862175e-07,
"logits/chosen": -0.45945605635643005,
"logits/rejected": -0.33354875445365906,
"logps/chosen": -71.20188903808594,
"logps/rejected": -212.95767211914062,
"loss": 0.5229,
"rewards/accuracies": 1.0,
"rewards/chosen": 0.16831260919570923,
"rewards/margins": 0.38181251287460327,
"rewards/rejected": -0.21349990367889404,
"step": 30
},
{
"epoch": 0.3611738148984199,
"eval_logits/chosen": -0.4618959426879883,
"eval_logits/rejected": -0.3433874249458313,
"eval_logps/chosen": -98.28858947753906,
"eval_logps/rejected": -212.0100555419922,
"eval_loss": 0.5059286952018738,
"eval_rewards/accuracies": 1.0,
"eval_rewards/chosen": 0.18195854127407074,
"eval_rewards/margins": 0.4181654751300812,
"eval_rewards/rejected": -0.23620688915252686,
"eval_runtime": 179.4182,
"eval_samples_per_second": 3.043,
"eval_steps_per_second": 1.522,
"step": 30
},
{
"epoch": 0.4213694507148232,
"grad_norm": 9.376220703125,
"learning_rate": 4.6692202414695724e-07,
"logits/chosen": -0.4632042944431305,
"logits/rejected": -0.35029542446136475,
"logps/chosen": -84.06396484375,
"logps/rejected": -213.848388671875,
"loss": 0.4976,
"rewards/accuracies": 1.0,
"rewards/chosen": 0.2034817487001419,
"rewards/margins": 0.4632874131202698,
"rewards/rejected": -0.25980567932128906,
"step": 35
},
{
"epoch": 0.4815650865312265,
"grad_norm": 8.679346084594727,
"learning_rate": 4.534239241377266e-07,
"logits/chosen": -0.44362330436706543,
"logits/rejected": -0.2992916703224182,
"logps/chosen": -105.2283706665039,
"logps/rejected": -244.84890747070312,
"loss": 0.4197,
"rewards/accuracies": 1.0,
"rewards/chosen": 0.22778573632240295,
"rewards/margins": 0.6910415291786194,
"rewards/rejected": -0.46325573325157166,
"step": 40
},
{
"epoch": 0.5417607223476298,
"grad_norm": 7.219143867492676,
"learning_rate": 4.3789121884703727e-07,
"logits/chosen": -0.41270333528518677,
"logits/rejected": -0.27924439311027527,
"logps/chosen": -70.08865356445312,
"logps/rejected": -261.56170654296875,
"loss": 0.3621,
"rewards/accuracies": 1.0,
"rewards/chosen": 0.23598209023475647,
"rewards/margins": 0.9187321662902832,
"rewards/rejected": -0.6827500462532043,
"step": 45
},
{
"epoch": 0.6019563581640331,
"grad_norm": 6.640863418579102,
"learning_rate": 4.204792632772754e-07,
"logits/chosen": -0.4174782633781433,
"logits/rejected": -0.2659801244735718,
"logps/chosen": -109.1211166381836,
"logps/rejected": -280.77813720703125,
"loss": 0.3123,
"rewards/accuracies": 1.0,
"rewards/chosen": 0.2913265824317932,
"rewards/margins": 1.1760694980621338,
"rewards/rejected": -0.8847430348396301,
"step": 50
}
],
"logging_steps": 5,
"max_steps": 166,
"num_input_tokens_seen": 0,
"num_train_epochs": 2,
"save_steps": 50,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 0.0,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}
|