| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 11080, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.18050541516245489, | |
| "grad_norm": 31.555688858032227, | |
| "learning_rate": 2.864620938628159e-05, | |
| "loss": 1.3005, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.36101083032490977, | |
| "grad_norm": 20.2050838470459, | |
| "learning_rate": 2.729241877256318e-05, | |
| "loss": 0.9978, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.5415162454873647, | |
| "grad_norm": 25.661306381225586, | |
| "learning_rate": 2.5938628158844765e-05, | |
| "loss": 0.9259, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.7220216606498195, | |
| "grad_norm": 22.348859786987305, | |
| "learning_rate": 2.4584837545126353e-05, | |
| "loss": 0.8798, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.9025270758122743, | |
| "grad_norm": 32.197166442871094, | |
| "learning_rate": 2.3231046931407943e-05, | |
| "loss": 0.8471, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.0830324909747293, | |
| "grad_norm": 20.764020919799805, | |
| "learning_rate": 2.1877256317689534e-05, | |
| "loss": 0.771, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.263537906137184, | |
| "grad_norm": 15.258258819580078, | |
| "learning_rate": 2.0523465703971117e-05, | |
| "loss": 0.6937, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.444043321299639, | |
| "grad_norm": 24.816614151000977, | |
| "learning_rate": 1.9169675090252708e-05, | |
| "loss": 0.709, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.6245487364620939, | |
| "grad_norm": 38.912071228027344, | |
| "learning_rate": 1.7815884476534298e-05, | |
| "loss": 0.6831, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.8050541516245486, | |
| "grad_norm": 8.910807609558105, | |
| "learning_rate": 1.6462093862815885e-05, | |
| "loss": 0.6742, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.9855595667870036, | |
| "grad_norm": 11.664189338684082, | |
| "learning_rate": 1.5108303249097474e-05, | |
| "loss": 0.69, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.1660649819494586, | |
| "grad_norm": 10.968308448791504, | |
| "learning_rate": 1.3754512635379063e-05, | |
| "loss": 0.5436, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.3465703971119134, | |
| "grad_norm": 11.711438179016113, | |
| "learning_rate": 1.240072202166065e-05, | |
| "loss": 0.5357, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.527075812274368, | |
| "grad_norm": 13.477335929870605, | |
| "learning_rate": 1.1046931407942239e-05, | |
| "loss": 0.5359, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.707581227436823, | |
| "grad_norm": 10.649256706237793, | |
| "learning_rate": 9.693140794223826e-06, | |
| "loss": 0.5394, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.888086642599278, | |
| "grad_norm": 10.525208473205566, | |
| "learning_rate": 8.339350180505416e-06, | |
| "loss": 0.5254, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.068592057761733, | |
| "grad_norm": 19.402320861816406, | |
| "learning_rate": 6.985559566787004e-06, | |
| "loss": 0.4775, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 3.2490974729241877, | |
| "grad_norm": 41.23615646362305, | |
| "learning_rate": 5.631768953068592e-06, | |
| "loss": 0.4003, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.4296028880866425, | |
| "grad_norm": 21.56231689453125, | |
| "learning_rate": 4.277978339350181e-06, | |
| "loss": 0.3952, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 3.6101083032490973, | |
| "grad_norm": 11.254490852355957, | |
| "learning_rate": 2.924187725631769e-06, | |
| "loss": 0.4007, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.7906137184115525, | |
| "grad_norm": 29.451414108276367, | |
| "learning_rate": 1.5703971119133576e-06, | |
| "loss": 0.3962, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 3.9711191335740073, | |
| "grad_norm": 16.022735595703125, | |
| "learning_rate": 2.1660649819494586e-07, | |
| "loss": 0.3853, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 11080, | |
| "total_flos": 1.0399493167607808e+16, | |
| "train_loss": 0.6484355885199261, | |
| "train_runtime": 8433.1347, | |
| "train_samples_per_second": 42.043, | |
| "train_steps_per_second": 1.314 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 11080, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "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": 1.0399493167607808e+16, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |