| { | |
| "best_metric": 0.9710213541984558, | |
| "best_model_checkpoint": "./outputs/checkpoint-1300", | |
| "epoch": 1.7333333333333334, | |
| "eval_steps": 100, | |
| "global_step": 1300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.0002, | |
| "loss": 1.001, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "eval_loss": 1.0910990238189697, | |
| "eval_runtime": 728.6528, | |
| "eval_samples_per_second": 2.649, | |
| "eval_steps_per_second": 0.332, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8971, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "eval_loss": 1.0677164793014526, | |
| "eval_runtime": 727.2702, | |
| "eval_samples_per_second": 2.654, | |
| "eval_steps_per_second": 0.333, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8815, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 1.052659273147583, | |
| "eval_runtime": 730.164, | |
| "eval_samples_per_second": 2.643, | |
| "eval_steps_per_second": 0.331, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8695, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "eval_loss": 1.0382879972457886, | |
| "eval_runtime": 732.8796, | |
| "eval_samples_per_second": 2.633, | |
| "eval_steps_per_second": 0.33, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8548, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "eval_loss": 1.0260452032089233, | |
| "eval_runtime": 730.1971, | |
| "eval_samples_per_second": 2.643, | |
| "eval_steps_per_second": 0.331, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8496, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 1.0152662992477417, | |
| "eval_runtime": 732.4802, | |
| "eval_samples_per_second": 2.635, | |
| "eval_steps_per_second": 0.33, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8445, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "eval_loss": 1.0105178356170654, | |
| "eval_runtime": 802.0691, | |
| "eval_samples_per_second": 2.406, | |
| "eval_steps_per_second": 0.302, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8346, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "eval_loss": 1.0066964626312256, | |
| "eval_runtime": 789.7964, | |
| "eval_samples_per_second": 2.444, | |
| "eval_steps_per_second": 0.306, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8199, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_loss": 0.9989904165267944, | |
| "eval_runtime": 786.9058, | |
| "eval_samples_per_second": 2.453, | |
| "eval_steps_per_second": 0.308, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8133, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "eval_loss": 0.9895688891410828, | |
| "eval_runtime": 782.3169, | |
| "eval_samples_per_second": 2.467, | |
| "eval_steps_per_second": 0.309, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8193, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "eval_loss": 0.9853964447975159, | |
| "eval_runtime": 777.984, | |
| "eval_samples_per_second": 2.481, | |
| "eval_steps_per_second": 0.311, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8091, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 0.9784607887268066, | |
| "eval_runtime": 780.7522, | |
| "eval_samples_per_second": 2.472, | |
| "eval_steps_per_second": 0.31, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7983, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "eval_loss": 0.9710213541984558, | |
| "eval_runtime": 787.7479, | |
| "eval_samples_per_second": 2.45, | |
| "eval_steps_per_second": 0.307, | |
| "step": 1300 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 3750, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 100, | |
| "total_flos": 3.3683016270053376e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |