| { | |
| "best_metric": 2.444390296936035, | |
| "best_model_checkpoint": "./outputs/checkpoint-700", | |
| "epoch": 0.5100182149362478, | |
| "eval_steps": 100, | |
| "global_step": 700, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0002, | |
| "loss": 2.7406, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_loss": 2.6424405574798584, | |
| "eval_runtime": 206.8728, | |
| "eval_samples_per_second": 30.328, | |
| "eval_steps_per_second": 3.795, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0002, | |
| "loss": 2.6056, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 2.591430187225342, | |
| "eval_runtime": 206.9647, | |
| "eval_samples_per_second": 30.314, | |
| "eval_steps_per_second": 3.793, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0002, | |
| "loss": 2.5631, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_loss": 2.5565075874328613, | |
| "eval_runtime": 206.6263, | |
| "eval_samples_per_second": 30.364, | |
| "eval_steps_per_second": 3.799, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0002, | |
| "loss": 2.5363, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 2.5222103595733643, | |
| "eval_runtime": 234.7988, | |
| "eval_samples_per_second": 26.721, | |
| "eval_steps_per_second": 3.343, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4935, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_loss": 2.4934604167938232, | |
| "eval_runtime": 206.7039, | |
| "eval_samples_per_second": 30.353, | |
| "eval_steps_per_second": 3.798, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0002, | |
| "loss": 2.467, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 2.467857837677002, | |
| "eval_runtime": 287.8992, | |
| "eval_samples_per_second": 21.792, | |
| "eval_steps_per_second": 2.727, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4494, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 2.444390296936035, | |
| "eval_runtime": 206.5306, | |
| "eval_samples_per_second": 30.378, | |
| "eval_steps_per_second": 3.801, | |
| "step": 700 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 4116, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 2.042513200078848e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |