darthludious/adaption_preference_trace_decisions

1{2  "best_global_step": null,3  "best_metric": null,4  "best_model_checkpoint": null,5  "epoch": 1.0,6  "eval_steps": 4,7  "global_step": 21,8  "is_hyper_param_search": false,9  "is_local_process_zero": true,10  "is_world_process_zero": true,11  "log_history": [12    {13      "epoch": 0.047619047619047616,14      "grad_norm": 0.43169543147087097,15      "learning_rate": 0.0,16      "loss": 1.51171875,17      "step": 118    },19    {20      "epoch": 0.09523809523809523,21      "grad_norm": 0.410955548286438,22      "learning_rate": 5e-05,23      "loss": 1.579833984375,24      "step": 225    },26    {27      "epoch": 0.14285714285714285,28      "grad_norm": 0.6624472141265869,29      "learning_rate": 4.97229876633906e-05,30      "loss": 1.60009765625,31      "step": 332    },33    {34      "epoch": 0.19047619047619047,35      "grad_norm": 0.4703623056411743,36      "learning_rate": 4.889877161664096e-05,37      "loss": 1.463623046875,38      "step": 439    },40    {41      "epoch": 0.23809523809523808,42      "grad_norm": 0.35136279463768005,43      "learning_rate": 4.7547646794238276e-05,44      "loss": 1.486328125,45      "step": 546    },47    {48      "epoch": 0.23809523809523808,49      "eval_loss": 1.4140625,50      "eval_runtime": 5.1445,51      "eval_samples_per_second": 0.194,52      "eval_steps_per_second": 0.194,53      "step": 554    },55    {56      "epoch": 0.2857142857142857,57      "grad_norm": 0.40738654136657715,58      "learning_rate": 4.570288237343632e-05,59      "loss": 1.33544921875,60      "step": 661    },62    {63      "epoch": 0.3333333333333333,64      "grad_norm": 0.4430812895298004,65      "learning_rate": 4.340990257669732e-05,66      "loss": 1.56640625,67      "step": 768    },69    {70      "epoch": 0.38095238095238093,71      "grad_norm": 0.4507165253162384,72      "learning_rate": 4.072516817658065e-05,73      "loss": 1.44091796875,74      "step": 875    },76    {77      "epoch": 0.42857142857142855,78      "grad_norm": 0.5336745977401733,79      "learning_rate": 3.771478624413981e-05,80      "loss": 1.314697265625,81      "step": 982    },83    {84      "epoch": 0.42857142857142855,85      "eval_loss": 1.328125,86      "eval_runtime": 5.0725,87      "eval_samples_per_second": 0.197,88      "eval_steps_per_second": 0.197,89      "step": 990    },91    {92      "epoch": 0.47619047619047616,93      "grad_norm": 0.3681015372276306,94      "learning_rate": 3.4452882373436316e-05,95      "loss": 1.433349609375,96      "step": 1097    },98    {99      "epoch": 0.5238095238095238,100      "grad_norm": 0.3549262583255768,101      "learning_rate": 3.1019775463405195e-05,102      "loss": 1.31689453125,103      "step": 11104    },105    {106      "epoch": 0.5714285714285714,107      "grad_norm": 0.2255212664604187,108      "learning_rate": 2.7500000000000004e-05,109      "loss": 0.7911376953125,110      "step": 12111    },112    {113      "epoch": 0.6190476190476191,114      "grad_norm": 0.36084261536598206,115      "learning_rate": 2.3980224536594803e-05,116      "loss": 1.3671875,117      "step": 13118    },119    {120      "epoch": 0.6190476190476191,121      "eval_loss": 1.28125,122      "eval_runtime": 5.0728,123      "eval_samples_per_second": 0.197,124      "eval_steps_per_second": 0.197,125      "step": 13126    },127    {128      "epoch": 0.6666666666666666,129      "grad_norm": 0.32899197936058044,130      "learning_rate": 2.054711762656369e-05,131      "loss": 1.302734375,132      "step": 14133    },134    {135      "epoch": 0.7142857142857143,136      "grad_norm": 0.34358394145965576,137      "learning_rate": 1.72852137558602e-05,138      "loss": 1.3017578125,139      "step": 15140    },141    {142      "epoch": 0.7619047619047619,143      "grad_norm": 0.2757423520088196,144      "learning_rate": 1.427483182341936e-05,145      "loss": 1.3701171875,146      "step": 16147    },148    {149      "epoch": 0.8095238095238095,150      "grad_norm": 0.3158814311027527,151      "learning_rate": 1.1590097423302684e-05,152      "loss": 1.449951171875,153      "step": 17154    },155    {156      "epoch": 0.8095238095238095,157      "eval_loss": 1.2578125,158      "eval_runtime": 5.0981,159      "eval_samples_per_second": 0.196,160      "eval_steps_per_second": 0.196,161      "step": 17162    },163    {164      "epoch": 0.8571428571428571,165      "grad_norm": 0.36973994970321655,166      "learning_rate": 9.297117626563687e-06,167      "loss": 1.319580078125,168      "step": 18169    },170    {171      "epoch": 0.9047619047619048,172      "grad_norm": 0.3462851941585541,173      "learning_rate": 7.452353205761725e-06,174      "loss": 1.38818359375,175      "step": 19176    },177    {178      "epoch": 0.9523809523809523,179      "grad_norm": 0.34591546654701233,180      "learning_rate": 6.1012283833590465e-06,181      "loss": 1.3006591796875,182      "step": 20183    },184    {185      "epoch": 1.0,186      "grad_norm": 0.2838385999202728,187      "learning_rate": 5.277012336609403e-06,188      "loss": 1.298583984375,189      "step": 21190    },191    {192      "epoch": 1.0,193      "eval_loss": 1.2578125,194      "eval_runtime": 5.0709,195      "eval_samples_per_second": 0.197,196      "eval_steps_per_second": 0.197,197      "step": 21198    }199  ],200  "logging_steps": 1.0,201  "max_steps": 21,202  "num_input_tokens_seen": 0,203  "num_train_epochs": 1,204  "save_steps": 0,205  "stateful_callbacks": {206    "TrainerControl": {207      "args": {208        "should_epoch_stop": false,209        "should_evaluate": false,210        "should_log": false,211        "should_save": true,212        "should_training_stop": true213      },214      "attributes": {}215    }216  },217  "total_flos": 1.0123420506469171e+17,218  "train_batch_size": 1,219  "trial_name": null,220  "trial_params": null221}222