| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.2518891687657431, | |
| "eval_steps": 3, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.010075566750629723, | |
| "grad_norm": 0.11719062924385071, | |
| "learning_rate": 2e-05, | |
| "loss": 11.7756, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.010075566750629723, | |
| "eval_loss": 11.775632858276367, | |
| "eval_runtime": 1.4254, | |
| "eval_samples_per_second": 58.929, | |
| "eval_steps_per_second": 29.464, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.020151133501259445, | |
| "grad_norm": 0.14791575074195862, | |
| "learning_rate": 4e-05, | |
| "loss": 11.7752, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.030226700251889168, | |
| "grad_norm": 0.14653781056404114, | |
| "learning_rate": 6e-05, | |
| "loss": 11.7731, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.030226700251889168, | |
| "eval_loss": 11.775500297546387, | |
| "eval_runtime": 1.4206, | |
| "eval_samples_per_second": 59.13, | |
| "eval_steps_per_second": 29.565, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.04030226700251889, | |
| "grad_norm": 0.15554942190647125, | |
| "learning_rate": 8e-05, | |
| "loss": 11.7685, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.05037783375314862, | |
| "grad_norm": 0.1331409513950348, | |
| "learning_rate": 0.0001, | |
| "loss": 11.7752, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.060453400503778336, | |
| "grad_norm": 0.17101934552192688, | |
| "learning_rate": 0.00012, | |
| "loss": 11.7782, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.060453400503778336, | |
| "eval_loss": 11.775053024291992, | |
| "eval_runtime": 1.4347, | |
| "eval_samples_per_second": 58.549, | |
| "eval_steps_per_second": 29.275, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.07052896725440806, | |
| "grad_norm": 0.11344543099403381, | |
| "learning_rate": 0.00014, | |
| "loss": 11.7742, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.08060453400503778, | |
| "grad_norm": 0.12048022449016571, | |
| "learning_rate": 0.00016, | |
| "loss": 11.7692, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0906801007556675, | |
| "grad_norm": 0.1138000637292862, | |
| "learning_rate": 0.00018, | |
| "loss": 11.7727, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0906801007556675, | |
| "eval_loss": 11.774245262145996, | |
| "eval_runtime": 1.4321, | |
| "eval_samples_per_second": 58.657, | |
| "eval_steps_per_second": 29.328, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.10075566750629723, | |
| "grad_norm": 0.11828505247831345, | |
| "learning_rate": 0.0002, | |
| "loss": 11.7817, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.11083123425692695, | |
| "grad_norm": 0.11759434640407562, | |
| "learning_rate": 0.00019781476007338058, | |
| "loss": 11.7652, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.12090680100755667, | |
| "grad_norm": 0.1497734636068344, | |
| "learning_rate": 0.0001913545457642601, | |
| "loss": 11.7804, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.12090680100755667, | |
| "eval_loss": 11.773073196411133, | |
| "eval_runtime": 1.4475, | |
| "eval_samples_per_second": 58.032, | |
| "eval_steps_per_second": 29.016, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.1309823677581864, | |
| "grad_norm": 0.12857705354690552, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 11.7778, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.14105793450881612, | |
| "grad_norm": 0.12967197597026825, | |
| "learning_rate": 0.00016691306063588583, | |
| "loss": 11.774, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.15113350125944586, | |
| "grad_norm": 0.12622714042663574, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 11.7798, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.15113350125944586, | |
| "eval_loss": 11.771885871887207, | |
| "eval_runtime": 1.4339, | |
| "eval_samples_per_second": 58.582, | |
| "eval_steps_per_second": 29.291, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.16120906801007556, | |
| "grad_norm": 0.0886993408203125, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 11.7713, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.1712846347607053, | |
| "grad_norm": 0.12222450226545334, | |
| "learning_rate": 0.00011045284632676536, | |
| "loss": 11.7706, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.181360201511335, | |
| "grad_norm": 0.1096186414361, | |
| "learning_rate": 8.954715367323468e-05, | |
| "loss": 11.7758, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.181360201511335, | |
| "eval_loss": 11.771020889282227, | |
| "eval_runtime": 1.4359, | |
| "eval_samples_per_second": 58.498, | |
| "eval_steps_per_second": 29.249, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.19143576826196473, | |
| "grad_norm": 0.12593254446983337, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 11.7686, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.20151133501259447, | |
| "grad_norm": 0.12455257028341293, | |
| "learning_rate": 5.000000000000002e-05, | |
| "loss": 11.7715, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.21158690176322417, | |
| "grad_norm": 0.13320545852184296, | |
| "learning_rate": 3.308693936411421e-05, | |
| "loss": 11.7673, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.21158690176322417, | |
| "eval_loss": 11.770564079284668, | |
| "eval_runtime": 1.4506, | |
| "eval_samples_per_second": 57.906, | |
| "eval_steps_per_second": 28.953, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.2216624685138539, | |
| "grad_norm": 0.1167505532503128, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 11.7674, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.23173803526448364, | |
| "grad_norm": 0.11704544723033905, | |
| "learning_rate": 8.645454235739903e-06, | |
| "loss": 11.7747, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.24181360201511334, | |
| "grad_norm": 0.14804768562316895, | |
| "learning_rate": 2.1852399266194314e-06, | |
| "loss": 11.7682, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.24181360201511334, | |
| "eval_loss": 11.770421028137207, | |
| "eval_runtime": 1.4338, | |
| "eval_samples_per_second": 58.584, | |
| "eval_steps_per_second": 29.292, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.2518891687657431, | |
| "grad_norm": 0.1328144669532776, | |
| "learning_rate": 0.0, | |
| "loss": 11.7815, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 25, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 5, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 10185002188800.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |