| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.057191878753217046, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.001143837575064341, | |
| "eval_loss": 1.3305572271347046, | |
| "eval_runtime": 25.1559, | |
| "eval_samples_per_second": 14.669, | |
| "eval_steps_per_second": 7.354, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.01143837575064341, | |
| "grad_norm": 0.8353750109672546, | |
| "learning_rate": 0.00019396926207859084, | |
| "loss": 5.2627, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.01486988847583643, | |
| "eval_loss": 1.2488889694213867, | |
| "eval_runtime": 25.1448, | |
| "eval_samples_per_second": 14.675, | |
| "eval_steps_per_second": 7.357, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.02287675150128682, | |
| "grad_norm": 0.9006792902946472, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 5.0163, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.02973977695167286, | |
| "eval_loss": 1.1914223432540894, | |
| "eval_runtime": 25.1429, | |
| "eval_samples_per_second": 14.676, | |
| "eval_steps_per_second": 7.358, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.03431512725193023, | |
| "grad_norm": 2.2693514823913574, | |
| "learning_rate": 8.263518223330697e-05, | |
| "loss": 4.6876, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04460966542750929, | |
| "eval_loss": 1.1729326248168945, | |
| "eval_runtime": 25.1532, | |
| "eval_samples_per_second": 14.67, | |
| "eval_steps_per_second": 7.355, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.04575350300257364, | |
| "grad_norm": 1.1591876745224, | |
| "learning_rate": 2.339555568810221e-05, | |
| "loss": 4.6645, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.057191878753217046, | |
| "grad_norm": 1.2325793504714966, | |
| "learning_rate": 0.0, | |
| "loss": 4.5842, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "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": 1.6124617496199168e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |