| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0011843663639952626, | |
| "eval_steps": 3, | |
| "global_step": 9, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0001315962626661403, | |
| "grad_norm": NaN, | |
| "learning_rate": 2e-05, | |
| "loss": 5.5138, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0001315962626661403, | |
| "eval_loss": NaN, | |
| "eval_runtime": 100.4037, | |
| "eval_samples_per_second": 31.871, | |
| "eval_steps_per_second": 15.936, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0002631925253322806, | |
| "grad_norm": NaN, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.00039478878799842083, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 2.0479, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00039478878799842083, | |
| "eval_loss": NaN, | |
| "eval_runtime": 96.7623, | |
| "eval_samples_per_second": 33.071, | |
| "eval_steps_per_second": 16.535, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0005263850506645612, | |
| "grad_norm": NaN, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0006579813133307014, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 6.4504, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0007895775759968417, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00012, | |
| "loss": 0.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0007895775759968417, | |
| "eval_loss": NaN, | |
| "eval_runtime": 99.5233, | |
| "eval_samples_per_second": 32.153, | |
| "eval_steps_per_second": 16.077, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.000921173838662982, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00014, | |
| "loss": 1.1069, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0010527701013291224, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00016, | |
| "loss": 0.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0011843663639952626, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00018, | |
| "loss": 3.3776, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0011843663639952626, | |
| "eval_loss": NaN, | |
| "eval_runtime": 104.6866, | |
| "eval_samples_per_second": 30.567, | |
| "eval_steps_per_second": 15.284, | |
| "step": 9 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 10, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 3, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1043584568524800.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |