| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.06506180871828236, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0013012361743656475, | |
| "grad_norm": 6.549399375915527, | |
| "learning_rate": 2e-05, | |
| "loss": 5.8478, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0013012361743656475, | |
| "eval_loss": 5.85587739944458, | |
| "eval_runtime": 19.5633, | |
| "eval_samples_per_second": 66.195, | |
| "eval_steps_per_second": 8.281, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.002602472348731295, | |
| "grad_norm": 5.998199939727783, | |
| "learning_rate": 4e-05, | |
| "loss": 5.5929, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.003903708523096942, | |
| "grad_norm": 6.909670829772949, | |
| "learning_rate": 6e-05, | |
| "loss": 6.0899, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00520494469746259, | |
| "grad_norm": 5.808619976043701, | |
| "learning_rate": 8e-05, | |
| "loss": 5.6341, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.006506180871828237, | |
| "grad_norm": 6.830719947814941, | |
| "learning_rate": 0.0001, | |
| "loss": 5.4022, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.006506180871828237, | |
| "eval_loss": 5.209164619445801, | |
| "eval_runtime": 19.5573, | |
| "eval_samples_per_second": 66.216, | |
| "eval_steps_per_second": 8.283, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.007807417046193884, | |
| "grad_norm": 7.51235818862915, | |
| "learning_rate": 0.00012, | |
| "loss": 5.2535, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.009108653220559532, | |
| "grad_norm": 10.392406463623047, | |
| "learning_rate": 0.00014, | |
| "loss": 4.4599, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.01040988939492518, | |
| "grad_norm": 8.97028636932373, | |
| "learning_rate": 0.00016, | |
| "loss": 3.9407, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.011711125569290826, | |
| "grad_norm": 7.354924201965332, | |
| "learning_rate": 0.00018, | |
| "loss": 2.9676, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.013012361743656473, | |
| "grad_norm": 6.167208194732666, | |
| "learning_rate": 0.0002, | |
| "loss": 2.2725, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.013012361743656473, | |
| "eval_loss": 1.3080393075942993, | |
| "eval_runtime": 19.5559, | |
| "eval_samples_per_second": 66.22, | |
| "eval_steps_per_second": 8.284, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.014313597918022121, | |
| "grad_norm": 5.213840484619141, | |
| "learning_rate": 0.0001996917333733128, | |
| "loss": 1.3746, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.015614834092387769, | |
| "grad_norm": 4.336225509643555, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 0.83, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.016916070266753416, | |
| "grad_norm": 4.607591152191162, | |
| "learning_rate": 0.00019723699203976766, | |
| "loss": 0.5198, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.018217306441119064, | |
| "grad_norm": 4.71095609664917, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 0.5275, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.01951854261548471, | |
| "grad_norm": 3.6073062419891357, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 0.2262, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.01951854261548471, | |
| "eval_loss": 0.16385002434253693, | |
| "eval_runtime": 19.5579, | |
| "eval_samples_per_second": 66.214, | |
| "eval_steps_per_second": 8.283, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.02081977878985036, | |
| "grad_norm": 3.0078632831573486, | |
| "learning_rate": 0.0001891006524188368, | |
| "loss": 0.1252, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.022121014964216004, | |
| "grad_norm": 6.549544334411621, | |
| "learning_rate": 0.00018526401643540922, | |
| "loss": 0.1807, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.02342225113858165, | |
| "grad_norm": 2.745868682861328, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 0.0516, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0247234873129473, | |
| "grad_norm": 0.7366054654121399, | |
| "learning_rate": 0.0001760405965600031, | |
| "loss": 0.0241, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.026024723487312947, | |
| "grad_norm": 0.8767889738082886, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 0.0172, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.026024723487312947, | |
| "eval_loss": 0.01193653978407383, | |
| "eval_runtime": 19.5552, | |
| "eval_samples_per_second": 66.223, | |
| "eval_steps_per_second": 8.284, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.027325959661678594, | |
| "grad_norm": 0.9785062074661255, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 0.011, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.028627195836044242, | |
| "grad_norm": 0.13425800204277039, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 0.0031, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.02992843201040989, | |
| "grad_norm": 0.11440292745828629, | |
| "learning_rate": 0.0001522498564715949, | |
| "loss": 0.0023, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.031229668184775537, | |
| "grad_norm": 0.16419219970703125, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 0.0022, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.03253090435914118, | |
| "grad_norm": 0.27950215339660645, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 0.0034, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.03253090435914118, | |
| "eval_loss": 0.0018979853484779596, | |
| "eval_runtime": 19.5548, | |
| "eval_samples_per_second": 66.224, | |
| "eval_steps_per_second": 8.284, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.03383214053350683, | |
| "grad_norm": 0.05157754197716713, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 0.0007, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.03513337670787248, | |
| "grad_norm": 0.05505343899130821, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 0.001, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.03643461288223813, | |
| "grad_norm": 0.03492051735520363, | |
| "learning_rate": 0.0001156434465040231, | |
| "loss": 0.0005, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.03773584905660377, | |
| "grad_norm": 0.0410466231405735, | |
| "learning_rate": 0.0001078459095727845, | |
| "loss": 0.0007, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.03903708523096942, | |
| "grad_norm": 0.027093220502138138, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0005, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.03903708523096942, | |
| "eval_loss": 0.000504123920109123, | |
| "eval_runtime": 19.5547, | |
| "eval_samples_per_second": 66.225, | |
| "eval_steps_per_second": 8.284, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04033832140533507, | |
| "grad_norm": 0.01767433062195778, | |
| "learning_rate": 9.215409042721552e-05, | |
| "loss": 0.0004, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.04163955757970072, | |
| "grad_norm": 0.010012276470661163, | |
| "learning_rate": 8.435655349597689e-05, | |
| "loss": 0.0002, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.04294079375406636, | |
| "grad_norm": 0.04254867136478424, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 0.0005, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.04424202992843201, | |
| "grad_norm": 0.005319009069353342, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 0.0002, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.04554326610279766, | |
| "grad_norm": 0.015941577032208443, | |
| "learning_rate": 6.173165676349103e-05, | |
| "loss": 0.0002, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.04554326610279766, | |
| "eval_loss": 0.00025833846302703023, | |
| "eval_runtime": 19.5515, | |
| "eval_samples_per_second": 66.235, | |
| "eval_steps_per_second": 8.286, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.0468445022771633, | |
| "grad_norm": 0.10623931884765625, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 0.0012, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.048145738451528954, | |
| "grad_norm": 0.006040714215487242, | |
| "learning_rate": 4.7750143528405126e-05, | |
| "loss": 0.0002, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.0494469746258946, | |
| "grad_norm": 0.010647974908351898, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 0.0003, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.05074821080026025, | |
| "grad_norm": 0.003610523883253336, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 0.0001, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.05204944697462589, | |
| "grad_norm": 0.0037193503230810165, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 0.0002, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.05204944697462589, | |
| "eval_loss": 0.00016716058598831296, | |
| "eval_runtime": 19.5673, | |
| "eval_samples_per_second": 66.182, | |
| "eval_steps_per_second": 8.279, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.053350683148991544, | |
| "grad_norm": 0.00527266226708889, | |
| "learning_rate": 2.3959403439996907e-05, | |
| "loss": 0.0002, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.05465191932335719, | |
| "grad_norm": 0.005722023546695709, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 0.0002, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.05595315549772284, | |
| "grad_norm": 0.005843067076057196, | |
| "learning_rate": 1.4735983564590783e-05, | |
| "loss": 0.0002, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.057254391672088484, | |
| "grad_norm": 0.002936769975349307, | |
| "learning_rate": 1.0899347581163221e-05, | |
| "loss": 0.0001, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.05855562784645413, | |
| "grad_norm": 0.0033105313777923584, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 0.0001, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.05855562784645413, | |
| "eval_loss": 0.00015007508045528084, | |
| "eval_runtime": 19.5511, | |
| "eval_samples_per_second": 66.237, | |
| "eval_steps_per_second": 8.286, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.05985686402081978, | |
| "grad_norm": 0.0019124951213598251, | |
| "learning_rate": 4.8943483704846475e-06, | |
| "loss": 0.0001, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.06115810019518542, | |
| "grad_norm": 0.0029391671996563673, | |
| "learning_rate": 2.7630079602323442e-06, | |
| "loss": 0.0001, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.062459336369551074, | |
| "grad_norm": 0.002002465771511197, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 0.0001, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.06376057254391672, | |
| "grad_norm": 0.0030635118018835783, | |
| "learning_rate": 3.0826662668720364e-07, | |
| "loss": 0.0001, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.06506180871828236, | |
| "grad_norm": 0.0038037565536797047, | |
| "learning_rate": 0.0, | |
| "loss": 0.0001, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.06506180871828236, | |
| "eval_loss": 0.0001471501891501248, | |
| "eval_runtime": 19.5581, | |
| "eval_samples_per_second": 66.213, | |
| "eval_steps_per_second": 8.283, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 70, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.28019470188544e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |