| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 3.0508474576271185, |
| "eval_steps": 30, |
| "global_step": 180, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.51, |
| "grad_norm": 5.693178653717041, |
| "learning_rate": 2.0949720670391062e-05, |
| "loss": 1.7061, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.51, |
| "eval_loss": 1.036990761756897, |
| "eval_runtime": 99.4587, |
| "eval_samples_per_second": 0.483, |
| "eval_steps_per_second": 0.06, |
| "step": 30 |
| }, |
| { |
| "epoch": 1.02, |
| "grad_norm": 7.891976833343506, |
| "learning_rate": 1.675977653631285e-05, |
| "loss": 0.9556, |
| "step": 60 |
| }, |
| { |
| "epoch": 1.02, |
| "eval_loss": 0.5036168098449707, |
| "eval_runtime": 99.1132, |
| "eval_samples_per_second": 0.484, |
| "eval_steps_per_second": 0.061, |
| "step": 60 |
| }, |
| { |
| "epoch": 1.53, |
| "grad_norm": 10.309865951538086, |
| "learning_rate": 1.2569832402234637e-05, |
| "loss": 0.3647, |
| "step": 90 |
| }, |
| { |
| "epoch": 1.53, |
| "eval_loss": 0.27036499977111816, |
| "eval_runtime": 99.1803, |
| "eval_samples_per_second": 0.484, |
| "eval_steps_per_second": 0.06, |
| "step": 90 |
| }, |
| { |
| "epoch": 2.03, |
| "grad_norm": 8.009493827819824, |
| "learning_rate": 8.379888268156424e-06, |
| "loss": 0.2179, |
| "step": 120 |
| }, |
| { |
| "epoch": 2.03, |
| "eval_loss": 0.13816428184509277, |
| "eval_runtime": 99.5171, |
| "eval_samples_per_second": 0.482, |
| "eval_steps_per_second": 0.06, |
| "step": 120 |
| }, |
| { |
| "epoch": 2.54, |
| "grad_norm": 8.932345390319824, |
| "learning_rate": 4.189944134078212e-06, |
| "loss": 0.1175, |
| "step": 150 |
| }, |
| { |
| "epoch": 2.54, |
| "eval_loss": 0.08441048860549927, |
| "eval_runtime": 99.1696, |
| "eval_samples_per_second": 0.484, |
| "eval_steps_per_second": 0.061, |
| "step": 150 |
| }, |
| { |
| "epoch": 3.05, |
| "grad_norm": 3.603414535522461, |
| "learning_rate": 0.0, |
| "loss": 0.0756, |
| "step": 180 |
| }, |
| { |
| "epoch": 3.05, |
| "eval_loss": 0.07085774838924408, |
| "eval_runtime": 99.0507, |
| "eval_samples_per_second": 0.485, |
| "eval_steps_per_second": 0.061, |
| "step": 180 |
| } |
| ], |
| "logging_steps": 30, |
| "max_steps": 180, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 4, |
| "save_steps": 30, |
| "total_flos": 7891554341486592.0, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
