| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 14624, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 2.897428884026258e-05, | |
| "loss": 4.2775, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 2.7948577680525165e-05, | |
| "loss": 2.9304, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 2.6922866520787748e-05, | |
| "loss": 2.3775, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 2.589715536105033e-05, | |
| "loss": 2.0721, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 2.4871444201312912e-05, | |
| "loss": 1.8554, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 2.3845733041575492e-05, | |
| "loss": 1.7415, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 2.2820021881838072e-05, | |
| "loss": 1.7061, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 2.179431072210066e-05, | |
| "loss": 1.6338, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 2.076859956236324e-05, | |
| "loss": 1.5858, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 1.9742888402625823e-05, | |
| "loss": 1.5215, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 1.8717177242888403e-05, | |
| "loss": 1.5361, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 1.7691466083150983e-05, | |
| "loss": 1.4856, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 1.6665754923413567e-05, | |
| "loss": 1.511, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 1.564004376367615e-05, | |
| "loss": 1.4694, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 1.4614332603938733e-05, | |
| "loss": 1.4379, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 1.3588621444201313e-05, | |
| "loss": 1.3644, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 1.2562910284463895e-05, | |
| "loss": 1.3269, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 1.1537199124726478e-05, | |
| "loss": 1.3391, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 1.0511487964989058e-05, | |
| "loss": 1.3217, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 9.485776805251642e-06, | |
| "loss": 1.3278, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 8.460065645514224e-06, | |
| "loss": 1.3462, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 7.434354485776806e-06, | |
| "loss": 1.3083, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 6.408643326039388e-06, | |
| "loss": 1.293, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 5.3829321663019695e-06, | |
| "loss": 1.347, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 4.357221006564551e-06, | |
| "loss": 1.2353, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 3.3315098468271337e-06, | |
| "loss": 1.2679, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 2.3057986870897156e-06, | |
| "loss": 1.2682, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 1.2800875273522977e-06, | |
| "loss": 1.2788, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 2.543763676148797e-07, | |
| "loss": 1.2428, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 14624, | |
| "total_flos": 1916220838136832.0, | |
| "train_loss": 0.22557616807699726, | |
| "train_runtime": 278.4262, | |
| "train_samples_per_second": 630.25, | |
| "train_steps_per_second": 52.524 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 14624, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 500, | |
| "total_flos": 1916220838136832.0, | |
| "train_batch_size": 12, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |