| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 0.35, |
| "eval_steps": 100, |
| "global_step": 700, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.0125, |
| "grad_norm": 12.285691261291504, |
| "learning_rate": 2.5555555555555554e-05, |
| "loss": 19.1959, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.025, |
| "grad_norm": 1.0500800609588623, |
| "learning_rate": 5.333333333333333e-05, |
| "loss": 7.7392, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.0375, |
| "grad_norm": 0.8419748544692993, |
| "learning_rate": 8.111111111111112e-05, |
| "loss": 1.2479, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.05, |
| "grad_norm": 1.2464624643325806, |
| "learning_rate": 0.00010888888888888889, |
| "loss": 0.7071, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.05, |
| "eval_loss": 0.5045744180679321, |
| "eval_runtime": 690.6274, |
| "eval_samples_per_second": 1.448, |
| "eval_steps_per_second": 1.448, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.0625, |
| "grad_norm": 0.8873828053474426, |
| "learning_rate": 0.00013666666666666666, |
| "loss": 0.4035, |
| "step": 125 |
| }, |
| { |
| "epoch": 0.075, |
| "grad_norm": 0.9160810708999634, |
| "learning_rate": 0.00016444444444444444, |
| "loss": 0.2371, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.0875, |
| "grad_norm": 0.7777899503707886, |
| "learning_rate": 0.00019222222222222224, |
| "loss": 0.1825, |
| "step": 175 |
| }, |
| { |
| "epoch": 0.1, |
| "grad_norm": 0.9546622037887573, |
| "learning_rate": 0.0001999952797459453, |
| "loss": 0.1328, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.1, |
| "eval_loss": 0.12972435355186462, |
| "eval_runtime": 689.9886, |
| "eval_samples_per_second": 1.449, |
| "eval_steps_per_second": 1.449, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.1125, |
| "grad_norm": 0.6794782876968384, |
| "learning_rate": 0.00019997306349823693, |
| "loss": 0.1162, |
| "step": 225 |
| }, |
| { |
| "epoch": 0.125, |
| "grad_norm": 0.8062716126441956, |
| "learning_rate": 0.00019993264143226513, |
| "loss": 0.1048, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.1375, |
| "grad_norm": 0.8031811118125916, |
| "learning_rate": 0.00019987402090918067, |
| "loss": 0.0988, |
| "step": 275 |
| }, |
| { |
| "epoch": 0.15, |
| "grad_norm": 0.37140652537345886, |
| "learning_rate": 0.000199797212604205, |
| "loss": 0.0885, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.15, |
| "eval_loss": 0.09123973548412323, |
| "eval_runtime": 689.6116, |
| "eval_samples_per_second": 1.45, |
| "eval_steps_per_second": 1.45, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.1625, |
| "grad_norm": 0.3891146779060364, |
| "learning_rate": 0.0001997022305046862, |
| "loss": 0.0797, |
| "step": 325 |
| }, |
| { |
| "epoch": 0.175, |
| "grad_norm": 0.2549072504043579, |
| "learning_rate": 0.00019958909190755187, |
| "loss": 0.0795, |
| "step": 350 |
| }, |
| { |
| "epoch": 0.1875, |
| "grad_norm": 0.16435463726520538, |
| "learning_rate": 0.0001994578174161592, |
| "loss": 0.0774, |
| "step": 375 |
| }, |
| { |
| "epoch": 0.2, |
| "grad_norm": 0.9256812930107117, |
| "learning_rate": 0.00019930843093654305, |
| "loss": 0.0737, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.2, |
| "eval_loss": 0.0729941874742508, |
| "eval_runtime": 690.0692, |
| "eval_samples_per_second": 1.449, |
| "eval_steps_per_second": 1.449, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.2125, |
| "grad_norm": 0.2688423693180084, |
| "learning_rate": 0.00019914095967306223, |
| "loss": 0.0703, |
| "step": 425 |
| }, |
| { |
| "epoch": 0.225, |
| "grad_norm": 0.24806413054466248, |
| "learning_rate": 0.0001989554341234457, |
| "loss": 0.0661, |
| "step": 450 |
| }, |
| { |
| "epoch": 0.2375, |
| "grad_norm": 0.2767386734485626, |
| "learning_rate": 0.00019875188807323842, |
| "loss": 0.0675, |
| "step": 475 |
| }, |
| { |
| "epoch": 0.25, |
| "grad_norm": 0.120990090072155, |
| "learning_rate": 0.00019853035858964906, |
| "loss": 0.0666, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.25, |
| "eval_loss": 0.06402209401130676, |
| "eval_runtime": 690.2132, |
| "eval_samples_per_second": 1.449, |
| "eval_steps_per_second": 1.449, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.2625, |
| "grad_norm": 0.17512480914592743, |
| "learning_rate": 0.0001982908860147995, |
| "loss": 0.0634, |
| "step": 525 |
| }, |
| { |
| "epoch": 0.275, |
| "grad_norm": 0.21513459086418152, |
| "learning_rate": 0.00019803351395837848, |
| "loss": 0.0667, |
| "step": 550 |
| }, |
| { |
| "epoch": 0.2875, |
| "grad_norm": 0.18517647683620453, |
| "learning_rate": 0.00019775828928969975, |
| "loss": 0.0626, |
| "step": 575 |
| }, |
| { |
| "epoch": 0.3, |
| "grad_norm": 0.15308107435703278, |
| "learning_rate": 0.00019746526212916705, |
| "loss": 0.0603, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.3, |
| "eval_loss": 0.06020086258649826, |
| "eval_runtime": 690.5491, |
| "eval_samples_per_second": 1.448, |
| "eval_steps_per_second": 1.448, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.3125, |
| "grad_norm": 0.1161598339676857, |
| "learning_rate": 0.00019715448583914659, |
| "loss": 0.06, |
| "step": 625 |
| }, |
| { |
| "epoch": 0.325, |
| "grad_norm": 0.13704052567481995, |
| "learning_rate": 0.0001968260170142496, |
| "loss": 0.0611, |
| "step": 650 |
| }, |
| { |
| "epoch": 0.3375, |
| "grad_norm": 0.1627456545829773, |
| "learning_rate": 0.0001964799154710258, |
| "loss": 0.0595, |
| "step": 675 |
| }, |
| { |
| "epoch": 0.35, |
| "grad_norm": 0.14293187856674194, |
| "learning_rate": 0.0001961162442370707, |
| "loss": 0.0606, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.35, |
| "eval_loss": 0.05934726819396019, |
| "eval_runtime": 690.4657, |
| "eval_samples_per_second": 1.448, |
| "eval_steps_per_second": 1.448, |
| "step": 700 |
| } |
| ], |
| "logging_steps": 25, |
| "max_steps": 6000, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "save_steps": 100, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.42445094371328e+16, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
