| { | |
| "best_metric": 0.9801864624023438, | |
| "best_model_checkpoint": "./outputs/checkpoint-1600", | |
| "epoch": 2.1333333333333333, | |
| "eval_steps": 100, | |
| "global_step": 1600, | |
| "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": 1.0596, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "eval_loss": 1.0747358798980713, | |
| "eval_runtime": 354.4756, | |
| "eval_samples_per_second": 5.445, | |
| "eval_steps_per_second": 0.683, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.0002, | |
| "loss": 0.9173, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "eval_loss": 1.0527141094207764, | |
| "eval_runtime": 353.1278, | |
| "eval_samples_per_second": 5.465, | |
| "eval_steps_per_second": 0.685, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0002, | |
| "loss": 0.9029, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 1.0407249927520752, | |
| "eval_runtime": 353.5098, | |
| "eval_samples_per_second": 5.46, | |
| "eval_steps_per_second": 0.685, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8938, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "eval_loss": 1.033090591430664, | |
| "eval_runtime": 353.8818, | |
| "eval_samples_per_second": 5.454, | |
| "eval_steps_per_second": 0.684, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.0002, | |
| "loss": 0.881, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "eval_loss": 1.0268442630767822, | |
| "eval_runtime": 352.5818, | |
| "eval_samples_per_second": 5.474, | |
| "eval_steps_per_second": 0.686, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8815, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 1.0178813934326172, | |
| "eval_runtime": 353.5441, | |
| "eval_samples_per_second": 5.459, | |
| "eval_steps_per_second": 0.684, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8785, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "eval_loss": 1.0156869888305664, | |
| "eval_runtime": 353.6562, | |
| "eval_samples_per_second": 5.457, | |
| "eval_steps_per_second": 0.684, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 0.0002, | |
| "loss": 0.868, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "eval_loss": 1.0099247694015503, | |
| "eval_runtime": 354.1025, | |
| "eval_samples_per_second": 5.45, | |
| "eval_steps_per_second": 0.683, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8634, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_loss": 1.0071921348571777, | |
| "eval_runtime": 353.8229, | |
| "eval_samples_per_second": 5.455, | |
| "eval_steps_per_second": 0.684, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8555, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "eval_loss": 0.9995353817939758, | |
| "eval_runtime": 353.9933, | |
| "eval_samples_per_second": 5.452, | |
| "eval_steps_per_second": 0.684, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8581, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "eval_loss": 0.9986774921417236, | |
| "eval_runtime": 353.5619, | |
| "eval_samples_per_second": 5.459, | |
| "eval_steps_per_second": 0.684, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8584, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 0.9940086007118225, | |
| "eval_runtime": 354.063, | |
| "eval_samples_per_second": 5.451, | |
| "eval_steps_per_second": 0.683, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8518, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "eval_loss": 0.9896215796470642, | |
| "eval_runtime": 353.5881, | |
| "eval_samples_per_second": 5.458, | |
| "eval_steps_per_second": 0.684, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8457, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "eval_loss": 0.9863357543945312, | |
| "eval_runtime": 354.32, | |
| "eval_samples_per_second": 5.447, | |
| "eval_steps_per_second": 0.683, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.0002, | |
| "loss": 0.847, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.9830333590507507, | |
| "eval_runtime": 353.3265, | |
| "eval_samples_per_second": 5.462, | |
| "eval_steps_per_second": 0.685, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8372, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "eval_loss": 0.9801864624023438, | |
| "eval_runtime": 354.2045, | |
| "eval_samples_per_second": 5.449, | |
| "eval_steps_per_second": 0.683, | |
| "step": 1600 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 2250, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 4.523634737775329e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |