| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1261.8181818181818, | |
| "eval_steps": 500, | |
| "global_step": 3470, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 36.36363636363637, | |
| "grad_norm": 3.4013822078704834, | |
| "learning_rate": 0.00019994268257026118, | |
| "loss": 3.996, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 72.72727272727273, | |
| "grad_norm": 5.365049362182617, | |
| "learning_rate": 0.000199766047623841, | |
| "loss": 1.3298, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 109.0909090909091, | |
| "grad_norm": 11.665435791015625, | |
| "learning_rate": 0.00019946668751528744, | |
| "loss": 0.727, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 145.45454545454547, | |
| "grad_norm": 4.6548357009887695, | |
| "learning_rate": 0.0001990509463238309, | |
| "loss": 0.523, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 181.8181818181818, | |
| "grad_norm": 5.07234525680542, | |
| "learning_rate": 0.0001985109326154774, | |
| "loss": 0.3655, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 218.1818181818182, | |
| "grad_norm": 7.260560035705566, | |
| "learning_rate": 0.00019785089851017787, | |
| "loss": 0.273, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 254.54545454545453, | |
| "grad_norm": 4.73112154006958, | |
| "learning_rate": 0.00019707164815781908, | |
| "loss": 0.2198, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 290.90909090909093, | |
| "grad_norm": 0.689785361289978, | |
| "learning_rate": 0.00019617413095492114, | |
| "loss": 0.1656, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 327.27272727272725, | |
| "grad_norm": 3.2497429847717285, | |
| "learning_rate": 0.00019515944038794384, | |
| "loss": 0.1227, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 363.6363636363636, | |
| "grad_norm": 8.12833309173584, | |
| "learning_rate": 0.0001940288127010419, | |
| "loss": 0.0802, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 400.0, | |
| "grad_norm": 5.075193405151367, | |
| "learning_rate": 0.000192783625389892, | |
| "loss": 0.0617, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 436.3636363636364, | |
| "grad_norm": 2.8493974208831787, | |
| "learning_rate": 0.00019142539552342638, | |
| "loss": 0.0404, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 472.72727272727275, | |
| "grad_norm": 0.399513840675354, | |
| "learning_rate": 0.00018995577789551803, | |
| "loss": 0.0327, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 509.09090909090907, | |
| "grad_norm": 1.5895440578460693, | |
| "learning_rate": 0.00018837656300886937, | |
| "loss": 0.0355, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 545.4545454545455, | |
| "grad_norm": 0.0846463143825531, | |
| "learning_rate": 0.00018668967489356028, | |
| "loss": 0.0127, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 581.8181818181819, | |
| "grad_norm": 0.04372192919254303, | |
| "learning_rate": 0.00018489716876291415, | |
| "loss": 0.0203, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 618.1818181818181, | |
| "grad_norm": 0.09042660892009735, | |
| "learning_rate": 0.00018300122850953675, | |
| "loss": 0.0125, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 654.5454545454545, | |
| "grad_norm": 0.06565279513597488, | |
| "learning_rate": 0.00018100416404457961, | |
| "loss": 0.0124, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 690.9090909090909, | |
| "grad_norm": 0.4747157692909241, | |
| "learning_rate": 0.00017890840848346908, | |
| "loss": 0.0159, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 727.2727272727273, | |
| "grad_norm": 0.011907286942005157, | |
| "learning_rate": 0.00017671651518153, | |
| "loss": 0.0092, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 763.6363636363636, | |
| "grad_norm": 0.16101345419883728, | |
| "learning_rate": 0.0001744311546231154, | |
| "loss": 0.0085, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 800.0, | |
| "grad_norm": 0.4659230411052704, | |
| "learning_rate": 0.00017205511116803306, | |
| "loss": 0.0084, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 836.3636363636364, | |
| "grad_norm": 0.026558930054306984, | |
| "learning_rate": 0.00016959127965923142, | |
| "loss": 0.011, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 872.7272727272727, | |
| "grad_norm": 0.07431953400373459, | |
| "learning_rate": 0.0001670426618958799, | |
| "loss": 0.012, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 909.0909090909091, | |
| "grad_norm": 0.17360587418079376, | |
| "learning_rate": 0.00016441236297613866, | |
| "loss": 0.0102, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 945.4545454545455, | |
| "grad_norm": 0.031083501875400543, | |
| "learning_rate": 0.00016173105296858452, | |
| "loss": 0.0068, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 981.8181818181819, | |
| "grad_norm": 0.3104077875614166, | |
| "learning_rate": 0.00015894783631276567, | |
| "loss": 0.0053, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1018.1818181818181, | |
| "grad_norm": 0.002124856458976865, | |
| "learning_rate": 0.0001560928007986178, | |
| "loss": 0.0062, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1054.5454545454545, | |
| "grad_norm": 0.005405977368354797, | |
| "learning_rate": 0.0001531694248471304, | |
| "loss": 0.0033, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 1090.909090909091, | |
| "grad_norm": 0.008714040741324425, | |
| "learning_rate": 0.00015018127014158886, | |
| "loss": 0.0071, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1127.2727272727273, | |
| "grad_norm": 0.04300049692392349, | |
| "learning_rate": 0.00014713197728821183, | |
| "loss": 0.0072, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 1163.6363636363637, | |
| "grad_norm": 0.00944207701832056, | |
| "learning_rate": 0.00014402526138063373, | |
| "loss": 0.006, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 1200.0, | |
| "grad_norm": 0.004597234074026346, | |
| "learning_rate": 0.0001408967639298663, | |
| "loss": 0.0138, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 1236.3636363636363, | |
| "grad_norm": 0.07871837168931961, | |
| "learning_rate": 0.00013768710104121627, | |
| "loss": 0.0069, | |
| "step": 3400 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 9000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4500, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.0175353452509184e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |