| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 275, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.7857142857142857e-06, | |
| "loss": 2.5611, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 2.5326, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 4.970937357800635e-05, | |
| "loss": 2.2958, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.9636107683181763, | |
| "eval_runtime": 1.2207, | |
| "eval_samples_per_second": 18.842, | |
| "eval_steps_per_second": 4.915, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 4.7957734321334915e-05, | |
| "loss": 1.9545, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 4.4728512734909844e-05, | |
| "loss": 1.8763, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 4.022954469173914e-05, | |
| "loss": 1.8837, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.9193000793457031, | |
| "eval_runtime": 1.2166, | |
| "eval_samples_per_second": 18.905, | |
| "eval_steps_per_second": 4.932, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 3.475038819391789e-05, | |
| "loss": 1.7375, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 2.864368713521274e-05, | |
| "loss": 1.7612, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 2.2302474793696117e-05, | |
| "loss": 1.7174, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.9123048782348633, | |
| "eval_runtime": 1.2199, | |
| "eval_samples_per_second": 18.853, | |
| "eval_steps_per_second": 4.918, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "learning_rate": 1.613487782393661e-05, | |
| "loss": 1.6455, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "learning_rate": 1.0537848824653418e-05, | |
| "loss": 1.6657, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 5.871618079580327e-06, | |
| "loss": 1.6515, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.9245940446853638, | |
| "eval_runtime": 1.269, | |
| "eval_samples_per_second": 18.125, | |
| "eval_steps_per_second": 4.728, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "learning_rate": 2.4365087825904355e-06, | |
| "loss": 1.5988, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 4.73, | |
| "learning_rate": 4.536079419643374e-07, | |
| "loss": 1.6337, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 1.9287786483764648, | |
| "eval_runtime": 1.2179, | |
| "eval_samples_per_second": 18.886, | |
| "eval_steps_per_second": 4.927, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 275, | |
| "total_flos": 9506378566434816.0, | |
| "train_loss": 1.8268643422560258, | |
| "train_runtime": 238.6826, | |
| "train_samples_per_second": 4.546, | |
| "train_steps_per_second": 1.152 | |
| } | |
| ], | |
| "logging_steps": 20, | |
| "max_steps": 275, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 20, | |
| "total_flos": 9506378566434816.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |