| { |
| "best_global_step": 5940, |
| "best_metric": 3.1114161014556885, |
| "best_model_checkpoint": null, |
| "epoch": 12.121212121212121, |
| "eval_steps": 500, |
| "global_step": 6000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.00202020202020202, |
| "grad_norm": 29.67241096496582, |
| "learning_rate": 0.0, |
| "loss": 10.3402, |
| "step": 1 |
| }, |
| { |
| "epoch": 1.0, |
| "grad_norm": 3.3088717460632324, |
| "learning_rate": 9.88e-05, |
| "loss": 4.8939, |
| "step": 495 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 3.929011583328247, |
| "eval_runtime": 29.0968, |
| "eval_samples_per_second": 1091.735, |
| "eval_steps_per_second": 4.296, |
| "step": 495 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 3.365527629852295, |
| "learning_rate": 9.478678038379531e-05, |
| "loss": 3.6468, |
| "step": 990 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 3.5377979278564453, |
| "eval_runtime": 29.5255, |
| "eval_samples_per_second": 1075.882, |
| "eval_steps_per_second": 4.234, |
| "step": 990 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 3.0418827533721924, |
| "learning_rate": 8.950959488272922e-05, |
| "loss": 3.3754, |
| "step": 1485 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 3.3858158588409424, |
| "eval_runtime": 29.5067, |
| "eval_samples_per_second": 1076.569, |
| "eval_steps_per_second": 4.236, |
| "step": 1485 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 3.1794700622558594, |
| "learning_rate": 8.42324093816631e-05, |
| "loss": 3.2248, |
| "step": 1980 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 3.2938942909240723, |
| "eval_runtime": 29.2967, |
| "eval_samples_per_second": 1084.287, |
| "eval_steps_per_second": 4.267, |
| "step": 1980 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 3.2430226802825928, |
| "learning_rate": 7.895522388059702e-05, |
| "loss": 3.1189, |
| "step": 2475 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_loss": 3.225740671157837, |
| "eval_runtime": 30.0345, |
| "eval_samples_per_second": 1057.65, |
| "eval_steps_per_second": 4.162, |
| "step": 2475 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 3.523634433746338, |
| "learning_rate": 7.367803837953093e-05, |
| "loss": 3.0357, |
| "step": 2970 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_loss": 3.1880686283111572, |
| "eval_runtime": 28.5036, |
| "eval_samples_per_second": 1114.456, |
| "eval_steps_per_second": 4.385, |
| "step": 2970 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 3.4692530632019043, |
| "learning_rate": 6.840085287846483e-05, |
| "loss": 2.9645, |
| "step": 3465 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_loss": 3.1580379009246826, |
| "eval_runtime": 29.6767, |
| "eval_samples_per_second": 1070.401, |
| "eval_steps_per_second": 4.212, |
| "step": 3465 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 3.635169744491577, |
| "learning_rate": 6.312366737739872e-05, |
| "loss": 2.9035, |
| "step": 3960 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_loss": 3.1353161334991455, |
| "eval_runtime": 29.4248, |
| "eval_samples_per_second": 1079.565, |
| "eval_steps_per_second": 4.248, |
| "step": 3960 |
| }, |
| { |
| "epoch": 9.0, |
| "grad_norm": 3.878169298171997, |
| "learning_rate": 5.784648187633263e-05, |
| "loss": 2.8468, |
| "step": 4455 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_loss": 3.1217947006225586, |
| "eval_runtime": 28.5843, |
| "eval_samples_per_second": 1111.308, |
| "eval_steps_per_second": 4.373, |
| "step": 4455 |
| }, |
| { |
| "epoch": 10.0, |
| "grad_norm": 4.254668712615967, |
| "learning_rate": 5.256929637526653e-05, |
| "loss": 2.7951, |
| "step": 4950 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_loss": 3.115081548690796, |
| "eval_runtime": 29.0237, |
| "eval_samples_per_second": 1094.483, |
| "eval_steps_per_second": 4.307, |
| "step": 4950 |
| }, |
| { |
| "epoch": 11.0, |
| "grad_norm": 4.000517845153809, |
| "learning_rate": 4.7292110874200426e-05, |
| "loss": 2.7449, |
| "step": 5445 |
| }, |
| { |
| "epoch": 11.0, |
| "eval_loss": 3.113570213317871, |
| "eval_runtime": 29.9952, |
| "eval_samples_per_second": 1059.035, |
| "eval_steps_per_second": 4.167, |
| "step": 5445 |
| }, |
| { |
| "epoch": 12.0, |
| "grad_norm": 4.319329261779785, |
| "learning_rate": 4.201492537313433e-05, |
| "loss": 2.6994, |
| "step": 5940 |
| }, |
| { |
| "epoch": 12.0, |
| "eval_loss": 3.1114161014556885, |
| "eval_runtime": 29.2008, |
| "eval_samples_per_second": 1087.848, |
| "eval_steps_per_second": 4.281, |
| "step": 5940 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 9900, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 20, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.00143306768384e+17, |
| "train_batch_size": 256, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
