| { | |
| "best_metric": 0.27896466851234436, | |
| "best_model_checkpoint": "out/checkpoint-900", | |
| "epoch": 0.8419083255378859, | |
| "global_step": 900, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09, | |
| "eval_loss": 0.3386548161506653, | |
| "eval_runtime": 3.4462, | |
| "eval_samples_per_second": 152.924, | |
| "eval_steps_per_second": 19.152, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "eval_loss": 0.32248201966285706, | |
| "eval_runtime": 3.2636, | |
| "eval_samples_per_second": 161.477, | |
| "eval_steps_per_second": 20.223, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "eval_loss": 0.3344081938266754, | |
| "eval_runtime": 3.3846, | |
| "eval_samples_per_second": 155.706, | |
| "eval_steps_per_second": 19.5, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "eval_loss": 0.37741541862487793, | |
| "eval_runtime": 3.4216, | |
| "eval_samples_per_second": 154.02, | |
| "eval_steps_per_second": 19.289, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 6.9799095266062985e-06, | |
| "loss": 0.3934, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "eval_loss": 0.41931670904159546, | |
| "eval_runtime": 3.2614, | |
| "eval_samples_per_second": 161.586, | |
| "eval_steps_per_second": 20.237, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "eval_loss": 0.3693557381629944, | |
| "eval_runtime": 3.2679, | |
| "eval_samples_per_second": 161.266, | |
| "eval_steps_per_second": 20.196, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "eval_loss": 0.34475621581077576, | |
| "eval_runtime": 3.2639, | |
| "eval_samples_per_second": 161.465, | |
| "eval_steps_per_second": 20.221, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_loss": 0.34064173698425293, | |
| "eval_runtime": 3.2698, | |
| "eval_samples_per_second": 161.174, | |
| "eval_steps_per_second": 20.185, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "eval_loss": 0.27896466851234436, | |
| "eval_runtime": 3.2598, | |
| "eval_samples_per_second": 161.668, | |
| "eval_steps_per_second": 20.247, | |
| "step": 900 | |
| } | |
| ], | |
| "max_steps": 2138, | |
| "num_train_epochs": 2, | |
| "total_flos": 148483879153632.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |