| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 8.695652173913043, |
| "eval_steps": 50, |
| "global_step": 200, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 2.17, |
| "learning_rate": 1.582608695652174e-05, |
| "loss": 0.3679, |
| "step": 50 |
| }, |
| { |
| "epoch": 2.17, |
| "eval_accuracy": 0.9938606027525791, |
| "eval_f1": 0.8871438638880499, |
| "eval_loss": 0.02303539402782917, |
| "eval_precision": 0.9611692844677138, |
| "eval_recall": 0.8237053654888764, |
| "eval_runtime": 4.8529, |
| "eval_samples_per_second": 125.285, |
| "eval_steps_per_second": 2.061, |
| "step": 50 |
| }, |
| { |
| "epoch": 4.35, |
| "learning_rate": 1.1478260869565218e-05, |
| "loss": 0.0136, |
| "step": 100 |
| }, |
| { |
| "epoch": 4.35, |
| "eval_accuracy": 0.9980242057352166, |
| "eval_f1": 0.9645270270270271, |
| "eval_loss": 0.009941783733665943, |
| "eval_precision": 0.9683436970039571, |
| "eval_recall": 0.9607403252944475, |
| "eval_runtime": 4.8458, |
| "eval_samples_per_second": 125.471, |
| "eval_steps_per_second": 2.064, |
| "step": 100 |
| }, |
| { |
| "epoch": 6.52, |
| "learning_rate": 7.130434782608696e-06, |
| "loss": 0.0069, |
| "step": 150 |
| }, |
| { |
| "epoch": 6.52, |
| "eval_accuracy": 0.998249753025717, |
| "eval_f1": 0.9647979139504564, |
| "eval_loss": 0.007377615664154291, |
| "eval_precision": 0.9612172944887735, |
| "eval_recall": 0.9684053094036269, |
| "eval_runtime": 4.9186, |
| "eval_samples_per_second": 123.611, |
| "eval_steps_per_second": 2.033, |
| "step": 150 |
| }, |
| { |
| "epoch": 8.7, |
| "learning_rate": 2.782608695652174e-06, |
| "loss": 0.005, |
| "step": 200 |
| }, |
| { |
| "epoch": 8.7, |
| "eval_accuracy": 0.9984527455871672, |
| "eval_f1": 0.9683325548808968, |
| "eval_loss": 0.006530961487442255, |
| "eval_precision": 0.9676997759522031, |
| "eval_recall": 0.9689661618994204, |
| "eval_runtime": 4.8158, |
| "eval_samples_per_second": 126.252, |
| "eval_steps_per_second": 2.077, |
| "step": 200 |
| } |
| ], |
| "logging_steps": 50, |
| "max_steps": 230, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 10, |
| "save_steps": 50, |
| "total_flos": 2882161195507584.0, |
| "train_batch_size": 64, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
