| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.11890606420927467, | |
| "eval_steps": 10, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0023781212841854932, | |
| "eval_loss": 2.7670791149139404, | |
| "eval_runtime": 5.2365, | |
| "eval_samples_per_second": 33.992, | |
| "eval_steps_per_second": 16.996, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.011890606420927468, | |
| "grad_norm": 25.248817443847656, | |
| "learning_rate": 5e-05, | |
| "loss": 11.273, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.023781212841854936, | |
| "grad_norm": 31.11591148376465, | |
| "learning_rate": 0.0001, | |
| "loss": 10.8987, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.023781212841854936, | |
| "eval_loss": 2.7416107654571533, | |
| "eval_runtime": 3.8611, | |
| "eval_samples_per_second": 46.101, | |
| "eval_steps_per_second": 23.051, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0356718192627824, | |
| "grad_norm": 31.424205780029297, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 10.7519, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.04756242568370987, | |
| "grad_norm": 19.314899444580078, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 11.9791, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.04756242568370987, | |
| "eval_loss": 2.6716339588165283, | |
| "eval_runtime": 3.9763, | |
| "eval_samples_per_second": 44.766, | |
| "eval_steps_per_second": 22.383, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.059453032104637336, | |
| "grad_norm": 45.63984298706055, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 10.7915, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0713436385255648, | |
| "grad_norm": 43.98304748535156, | |
| "learning_rate": 5e-05, | |
| "loss": 11.0255, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0713436385255648, | |
| "eval_loss": 2.6170427799224854, | |
| "eval_runtime": 3.9764, | |
| "eval_samples_per_second": 44.764, | |
| "eval_steps_per_second": 22.382, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.08323424494649227, | |
| "grad_norm": 52.070167541503906, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 10.3994, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.09512485136741974, | |
| "grad_norm": 28.898639678955078, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 10.6568, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.09512485136741974, | |
| "eval_loss": 2.6050915718078613, | |
| "eval_runtime": 3.9033, | |
| "eval_samples_per_second": 45.602, | |
| "eval_steps_per_second": 22.801, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.1070154577883472, | |
| "grad_norm": 32.772613525390625, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 10.5308, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.11890606420927467, | |
| "grad_norm": 31.561771392822266, | |
| "learning_rate": 0.0, | |
| "loss": 10.5122, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.11890606420927467, | |
| "eval_loss": 2.6030380725860596, | |
| "eval_runtime": 3.922, | |
| "eval_samples_per_second": 45.385, | |
| "eval_steps_per_second": 22.693, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 532951037116416.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |