| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.076923076923077, | |
| "eval_steps": 5, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15384615384615385, | |
| "grad_norm": 0.9869943189081838, | |
| "learning_rate": 0.0001, | |
| "loss": 1.674, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 2.264981292489023, | |
| "learning_rate": 0.00018660254037844388, | |
| "loss": 1.6335, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.7692307692307693, | |
| "eval_loss": 1.3072189092636108, | |
| "eval_runtime": 2.5711, | |
| "eval_samples_per_second": 7.779, | |
| "eval_steps_per_second": 1.167, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 0.7994859628506756, | |
| "learning_rate": 0.00011736481776669306, | |
| "loss": 0.9233, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "eval_loss": 1.0146267414093018, | |
| "eval_runtime": 1.2915, | |
| "eval_samples_per_second": 15.486, | |
| "eval_steps_per_second": 2.323, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": 0.7302665680016867, | |
| "learning_rate": 3.5721239031346066e-05, | |
| "loss": 0.7059, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "eval_loss": 0.9329476356506348, | |
| "eval_runtime": 1.2822, | |
| "eval_samples_per_second": 15.598, | |
| "eval_steps_per_second": 2.34, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 0.6740041947230365, | |
| "learning_rate": 0.0, | |
| "loss": 0.6532, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "eval_loss": 0.9111747741699219, | |
| "eval_runtime": 1.2917, | |
| "eval_samples_per_second": 15.484, | |
| "eval_steps_per_second": 2.323, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "step": 20, | |
| "total_flos": 2367492882432.0, | |
| "train_loss": 0.9810080707073212, | |
| "train_runtime": 85.2326, | |
| "train_samples_per_second": 3.754, | |
| "train_steps_per_second": 0.235 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 1000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2367492882432.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |