| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.04441913439635535, | |
| "eval_steps": 10, | |
| "global_step": 39, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0011389521640091116, | |
| "eval_loss": 7.603264808654785, | |
| "eval_runtime": 4.9081, | |
| "eval_samples_per_second": 75.386, | |
| "eval_steps_per_second": 37.693, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0056947608200455585, | |
| "grad_norm": 1.7203195095062256, | |
| "learning_rate": 5e-05, | |
| "loss": 7.5131, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.011389521640091117, | |
| "grad_norm": 1.6863157749176025, | |
| "learning_rate": 0.0001, | |
| "loss": 7.6617, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.011389521640091117, | |
| "eval_loss": 7.561071395874023, | |
| "eval_runtime": 4.9798, | |
| "eval_samples_per_second": 74.301, | |
| "eval_steps_per_second": 37.15, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.017084282460136675, | |
| "grad_norm": 1.7132402658462524, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 7.5253, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.022779043280182234, | |
| "grad_norm": 1.3183465003967285, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 7.3096, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.022779043280182234, | |
| "eval_loss": 7.4392523765563965, | |
| "eval_runtime": 5.0678, | |
| "eval_samples_per_second": 73.01, | |
| "eval_steps_per_second": 36.505, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.02847380410022779, | |
| "grad_norm": 1.1598849296569824, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 7.3182, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.03416856492027335, | |
| "grad_norm": 2.4453885555267334, | |
| "learning_rate": 5e-05, | |
| "loss": 7.4379, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.03416856492027335, | |
| "eval_loss": 7.355154037475586, | |
| "eval_runtime": 5.0816, | |
| "eval_samples_per_second": 72.812, | |
| "eval_steps_per_second": 36.406, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.03986332574031891, | |
| "grad_norm": 1.6234405040740967, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 7.5061, | |
| "step": 35 | |
| } | |
| ], | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4376032837632.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |