| { | |
| "best_metric": 0.873400365630713, | |
| "best_model_checkpoint": "berttiny-hate_speech18-bothpretrained\\run-12\\checkpoint-600", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 600, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 6.854342606336822e-05, | |
| "loss": 0.499, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8555758683729433, | |
| "eval_loss": 0.48665618896484375, | |
| "eval_runtime": 10.8301, | |
| "eval_samples_per_second": 202.03, | |
| "eval_steps_per_second": 1.662, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 5.140756954752616e-05, | |
| "loss": 0.4699, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8592321755027422, | |
| "eval_loss": 0.48114827275276184, | |
| "eval_runtime": 10.9276, | |
| "eval_samples_per_second": 200.227, | |
| "eval_steps_per_second": 1.647, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 3.427171303168411e-05, | |
| "loss": 0.4591, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8724862888482633, | |
| "eval_loss": 0.4719109833240509, | |
| "eval_runtime": 10.8837, | |
| "eval_samples_per_second": 201.035, | |
| "eval_steps_per_second": 1.654, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 1.7135856515842055e-05, | |
| "loss": 0.4513, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8720292504570384, | |
| "eval_loss": 0.4705061912536621, | |
| "eval_runtime": 10.8611, | |
| "eval_samples_per_second": 201.454, | |
| "eval_steps_per_second": 1.657, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.4479, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.873400365630713, | |
| "eval_loss": 0.46913647651672363, | |
| "eval_runtime": 10.7987, | |
| "eval_samples_per_second": 202.617, | |
| "eval_steps_per_second": 1.667, | |
| "step": 600 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 600, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 9921791149920.0, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 8.567928257921027e-05, | |
| "num_train_epochs": 5, | |
| "per_device_train_batch_size": 64, | |
| "seed": 10 | |
| } | |
| } | |