| { | |
| "best_metric": 0.5116071701049805, | |
| "best_model_checkpoint": "tweet_sentiment/checkpoint-1336", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 2004, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 1.70059880239521e-05, | |
| "loss": 0.6771, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.772573997751967, | |
| "eval_loss": 0.5403854250907898, | |
| "eval_runtime": 26.1276, | |
| "eval_samples_per_second": 102.152, | |
| "eval_steps_per_second": 6.392, | |
| "step": 668 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 1.4011976047904192e-05, | |
| "loss": 0.5016, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7871862120644436, | |
| "eval_loss": 0.5116071701049805, | |
| "eval_runtime": 26.3763, | |
| "eval_samples_per_second": 101.189, | |
| "eval_steps_per_second": 6.331, | |
| "step": 1336 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 1.1017964071856288e-05, | |
| "loss": 0.4198, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 8.023952095808385e-06, | |
| "loss": 0.3337, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7924316223304608, | |
| "eval_loss": 0.5666561722755432, | |
| "eval_runtime": 26.0935, | |
| "eval_samples_per_second": 102.286, | |
| "eval_steps_per_second": 6.4, | |
| "step": 2004 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 3340, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 2273027931010656.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |