| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 179.53636841073939, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 2.132, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 251.9635633879995, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 2.5698, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 152.8902496957368, | |
| "learning_rate": 6e-06, | |
| "loss": 1.768, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 39.306296179877485, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.9482, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 15.99491838072262, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7738, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 5.838903588486864, | |
| "learning_rate": 9.987820251299121e-06, | |
| "loss": 0.5098, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 4.783670137975404, | |
| "learning_rate": 9.951340343707852e-06, | |
| "loss": 0.3721, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 3.6109544656639025, | |
| "learning_rate": 9.890738003669029e-06, | |
| "loss": 0.2693, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 1.7838453759518462, | |
| "learning_rate": 9.806308479691595e-06, | |
| "loss": 0.2655, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 1.0136025622608915, | |
| "learning_rate": 9.698463103929542e-06, | |
| "loss": 0.2257, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 1.8942955529683139, | |
| "learning_rate": 9.567727288213005e-06, | |
| "loss": 0.2505, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 1.7740057720955722, | |
| "learning_rate": 9.414737964294636e-06, | |
| "loss": 0.2153, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 1.299424789487767, | |
| "learning_rate": 9.24024048078213e-06, | |
| "loss": 0.2015, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.8241827071039669, | |
| "learning_rate": 9.045084971874738e-06, | |
| "loss": 0.1962, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 1.2657759943266225, | |
| "learning_rate": 8.83022221559489e-06, | |
| "loss": 0.2104, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.6199251066045637, | |
| "learning_rate": 8.596699001693257e-06, | |
| "loss": 0.1858, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 1.0082588167041746, | |
| "learning_rate": 8.345653031794292e-06, | |
| "loss": 0.1962, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 1.3806733105910687, | |
| "learning_rate": 8.078307376628292e-06, | |
| "loss": 0.1922, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 0.723149823087172, | |
| "learning_rate": 7.795964517353734e-06, | |
| "loss": 0.1775, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.9173976871827653, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.1878, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 1.0453801125015292, | |
| "learning_rate": 7.191855733945388e-06, | |
| "loss": 0.1737, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 0.9248133544590185, | |
| "learning_rate": 6.873032967079562e-06, | |
| "loss": 0.184, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 1.5551843239370804, | |
| "learning_rate": 6.545084971874738e-06, | |
| "loss": 0.163, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.8847356518931627, | |
| "learning_rate": 6.209609477998339e-06, | |
| "loss": 0.1734, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.6812558639710236, | |
| "learning_rate": 5.8682408883346535e-06, | |
| "loss": 0.1569, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 1.262922291164637, | |
| "learning_rate": 5.522642316338268e-06, | |
| "loss": 0.1777, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 0.6617931583057793, | |
| "learning_rate": 5.174497483512506e-06, | |
| "loss": 0.1622, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.9677789314295031, | |
| "learning_rate": 4.825502516487497e-06, | |
| "loss": 0.1604, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 0.6102597975752135, | |
| "learning_rate": 4.477357683661734e-06, | |
| "loss": 0.1376, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.8360072005534209, | |
| "learning_rate": 4.131759111665349e-06, | |
| "loss": 0.1633, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 0.6629076631146715, | |
| "learning_rate": 3.790390522001662e-06, | |
| "loss": 0.1354, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.6314863269976039, | |
| "learning_rate": 3.4549150281252635e-06, | |
| "loss": 0.1548, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 0.6606730970709336, | |
| "learning_rate": 3.12696703292044e-06, | |
| "loss": 0.1508, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 0.7610654446520294, | |
| "learning_rate": 2.8081442660546126e-06, | |
| "loss": 0.1555, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 0.5915047659147039, | |
| "learning_rate": 2.5000000000000015e-06, | |
| "loss": 0.1414, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.5677743309663439, | |
| "learning_rate": 2.204035482646267e-06, | |
| "loss": 0.1367, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 0.548800059070961, | |
| "learning_rate": 1.9216926233717087e-06, | |
| "loss": 0.1164, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 0.5673284409321975, | |
| "learning_rate": 1.6543469682057105e-06, | |
| "loss": 0.1337, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 0.7602183062829035, | |
| "learning_rate": 1.4033009983067454e-06, | |
| "loss": 0.1501, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.5136603955006809, | |
| "learning_rate": 1.1697777844051105e-06, | |
| "loss": 0.1333, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 0.5425021614926135, | |
| "learning_rate": 9.549150281252633e-07, | |
| "loss": 0.1405, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 0.6257075274385691, | |
| "learning_rate": 7.597595192178702e-07, | |
| "loss": 0.1511, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 0.5431160370524597, | |
| "learning_rate": 5.852620357053651e-07, | |
| "loss": 0.1515, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.6088309676882525, | |
| "learning_rate": 4.322727117869951e-07, | |
| "loss": 0.1406, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 0.5000759583820149, | |
| "learning_rate": 3.015368960704584e-07, | |
| "loss": 0.1319, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 0.588422977399883, | |
| "learning_rate": 1.9369152030840553e-07, | |
| "loss": 0.1416, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 0.47388093466019365, | |
| "learning_rate": 1.0926199633097156e-07, | |
| "loss": 0.1414, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.47998421829050647, | |
| "learning_rate": 4.865965629214819e-08, | |
| "loss": 0.1411, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "grad_norm": 0.4988223499901082, | |
| "learning_rate": 1.2179748700879013e-08, | |
| "loss": 0.1212, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.502592566297319, | |
| "learning_rate": 0.0, | |
| "loss": 0.1352, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.12961728870868683, | |
| "eval_runtime": 7.4391, | |
| "eval_samples_per_second": 11.292, | |
| "eval_steps_per_second": 2.823, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 50, | |
| "total_flos": 11042438184960.0, | |
| "train_loss": 0.32609067142009734, | |
| "train_runtime": 919.7114, | |
| "train_samples_per_second": 1.718, | |
| "train_steps_per_second": 0.054 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 11042438184960.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |