| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.07742934572202866, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0007742934572202865, | |
| "eval_loss": 0.5842716097831726, | |
| "eval_runtime": 186.038, | |
| "eval_samples_per_second": 11.697, | |
| "eval_steps_per_second": 1.462, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0023228803716608595, | |
| "grad_norm": 14.163662910461426, | |
| "learning_rate": 3e-05, | |
| "loss": 1.8208, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.004645760743321719, | |
| "grad_norm": 9.894718170166016, | |
| "learning_rate": 6e-05, | |
| "loss": 1.4177, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.006968641114982578, | |
| "grad_norm": 48.34400177001953, | |
| "learning_rate": 9e-05, | |
| "loss": 0.7921, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.006968641114982578, | |
| "eval_loss": 0.1012289971113205, | |
| "eval_runtime": 188.0373, | |
| "eval_samples_per_second": 11.572, | |
| "eval_steps_per_second": 1.447, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.009291521486643438, | |
| "grad_norm": 5.306800842285156, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.2852, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.011614401858304297, | |
| "grad_norm": 3.0954625606536865, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 0.1949, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.013937282229965157, | |
| "grad_norm": 1.6799471378326416, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.0677, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.013937282229965157, | |
| "eval_loss": 0.010273167863488197, | |
| "eval_runtime": 188.1518, | |
| "eval_samples_per_second": 11.565, | |
| "eval_steps_per_second": 1.446, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.016260162601626018, | |
| "grad_norm": 5.278903961181641, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.0318, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.018583042973286876, | |
| "grad_norm": 1.8316158056259155, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.0377, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.020905923344947737, | |
| "grad_norm": 0.3559686243534088, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 0.0169, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.020905923344947737, | |
| "eval_loss": 0.006285559851676226, | |
| "eval_runtime": 188.0824, | |
| "eval_samples_per_second": 11.569, | |
| "eval_steps_per_second": 1.446, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.023228803716608595, | |
| "grad_norm": 0.6742556095123291, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.0237, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.025551684088269456, | |
| "grad_norm": 1.4266101121902466, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 0.0166, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.027874564459930314, | |
| "grad_norm": 1.627569556236267, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.0115, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.027874564459930314, | |
| "eval_loss": 0.0022949930280447006, | |
| "eval_runtime": 188.026, | |
| "eval_samples_per_second": 11.573, | |
| "eval_steps_per_second": 1.447, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.030197444831591175, | |
| "grad_norm": 2.386284112930298, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 0.005, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.032520325203252036, | |
| "grad_norm": 0.3715166449546814, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.0047, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.03484320557491289, | |
| "grad_norm": 0.12145501375198364, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.0009, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.03484320557491289, | |
| "eval_loss": 0.001899574650451541, | |
| "eval_runtime": 188.0378, | |
| "eval_samples_per_second": 11.572, | |
| "eval_steps_per_second": 1.447, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.03716608594657375, | |
| "grad_norm": 0.957979142665863, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.0077, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.03948896631823461, | |
| "grad_norm": 0.04404503479599953, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 0.007, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.041811846689895474, | |
| "grad_norm": 0.040287747979164124, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 0.0027, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.041811846689895474, | |
| "eval_loss": 0.0014874738408252597, | |
| "eval_runtime": 188.104, | |
| "eval_samples_per_second": 11.568, | |
| "eval_steps_per_second": 1.446, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.04413472706155633, | |
| "grad_norm": 0.13004916906356812, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 0.0022, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.04645760743321719, | |
| "grad_norm": 0.2595899999141693, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.0009, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.04878048780487805, | |
| "grad_norm": 2.1593716144561768, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 0.0074, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.04878048780487805, | |
| "eval_loss": 0.0013361426535993814, | |
| "eval_runtime": 188.094, | |
| "eval_samples_per_second": 11.569, | |
| "eval_steps_per_second": 1.446, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.05110336817653891, | |
| "grad_norm": 0.09437140822410583, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 0.0027, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.053426248548199766, | |
| "grad_norm": 0.028786564245820045, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 0.0012, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.05574912891986063, | |
| "grad_norm": 0.012601899914443493, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 0.0061, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.05574912891986063, | |
| "eval_loss": 0.0011122261639684439, | |
| "eval_runtime": 188.0667, | |
| "eval_samples_per_second": 11.57, | |
| "eval_steps_per_second": 1.446, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.05807200929152149, | |
| "grad_norm": 0.679633617401123, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 0.0067, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.06039488966318235, | |
| "grad_norm": 0.323636531829834, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 0.0032, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.0627177700348432, | |
| "grad_norm": 0.45798206329345703, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 0.0025, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.0627177700348432, | |
| "eval_loss": 0.0009968923404812813, | |
| "eval_runtime": 188.0647, | |
| "eval_samples_per_second": 11.57, | |
| "eval_steps_per_second": 1.446, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.06504065040650407, | |
| "grad_norm": 0.01450399775058031, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 0.0002, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.06736353077816493, | |
| "grad_norm": 0.11589731276035309, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 0.0032, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.06968641114982578, | |
| "grad_norm": 0.09107878804206848, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 0.0009, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.06968641114982578, | |
| "eval_loss": 0.0009773264173418283, | |
| "eval_runtime": 188.1234, | |
| "eval_samples_per_second": 11.567, | |
| "eval_steps_per_second": 1.446, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.07200929152148665, | |
| "grad_norm": 0.19633813202381134, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 0.0033, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.0743321718931475, | |
| "grad_norm": 0.05542326718568802, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 0.0012, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.07665505226480836, | |
| "grad_norm": 0.8960587978363037, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 0.0027, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.07665505226480836, | |
| "eval_loss": 0.0009747794247232378, | |
| "eval_runtime": 187.9934, | |
| "eval_samples_per_second": 11.575, | |
| "eval_steps_per_second": 1.447, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.885215813355438e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |