| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 50, | |
| "global_step": 222, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06756756756756757, | |
| "grad_norm": 18.160281746562354, | |
| "learning_rate": 5e-07, | |
| "loss": 1.725, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.13513513513513514, | |
| "grad_norm": 12.466524806615743, | |
| "learning_rate": 1e-06, | |
| "loss": 1.5906, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.20270270270270271, | |
| "grad_norm": 8.158179790073767, | |
| "learning_rate": 9.98628141419305e-07, | |
| "loss": 1.2918, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.2702702702702703, | |
| "grad_norm": 4.085200528685059, | |
| "learning_rate": 9.94520093661082e-07, | |
| "loss": 1.1519, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.33783783783783783, | |
| "grad_norm": 3.850134297695916, | |
| "learning_rate": 9.876983993675989e-07, | |
| "loss": 1.0558, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.40540540540540543, | |
| "grad_norm": 3.6261076115421855, | |
| "learning_rate": 9.78200492138261e-07, | |
| "loss": 0.998, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.47297297297297297, | |
| "grad_norm": 3.5652322410577693, | |
| "learning_rate": 9.66078491115194e-07, | |
| "loss": 0.9869, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.5405405405405406, | |
| "grad_norm": 3.429856478581444, | |
| "learning_rate": 9.513989149828717e-07, | |
| "loss": 0.9609, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.6081081081081081, | |
| "grad_norm": 3.503725889554788, | |
| "learning_rate": 9.342423169512071e-07, | |
| "loss": 0.9515, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.6756756756756757, | |
| "grad_norm": 3.9352337469583434, | |
| "learning_rate": 9.147028427251009e-07, | |
| "loss": 0.9353, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6756756756756757, | |
| "eval_loss": 0.9305333495140076, | |
| "eval_runtime": 36.7139, | |
| "eval_samples_per_second": 57.199, | |
| "eval_steps_per_second": 0.899, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.7432432432432432, | |
| "grad_norm": 3.3810767942983877, | |
| "learning_rate": 8.928877138860706e-07, | |
| "loss": 0.9362, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.8108108108108109, | |
| "grad_norm": 3.480739656688697, | |
| "learning_rate": 8.689166395208636e-07, | |
| "loss": 0.9287, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.8783783783783784, | |
| "grad_norm": 3.5875143505506037, | |
| "learning_rate": 8.429211593257052e-07, | |
| "loss": 0.94, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.9459459459459459, | |
| "grad_norm": 3.3991304127197157, | |
| "learning_rate": 8.150439217908556e-07, | |
| "loss": 0.8969, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.0135135135135136, | |
| "grad_norm": 3.739421526863416, | |
| "learning_rate": 7.854379014263876e-07, | |
| "loss": 0.8994, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.0810810810810811, | |
| "grad_norm": 3.315727217565176, | |
| "learning_rate": 7.542655593246103e-07, | |
| "loss": 0.8338, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.1486486486486487, | |
| "grad_norm": 3.481817716947705, | |
| "learning_rate": 7.216979516654943e-07, | |
| "loss": 0.8294, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.2162162162162162, | |
| "grad_norm": 3.4570894761347306, | |
| "learning_rate": 6.87913791057119e-07, | |
| "loss": 0.8202, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.2837837837837838, | |
| "grad_norm": 3.4227093804265443, | |
| "learning_rate": 6.530984658619733e-07, | |
| "loss": 0.8417, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 1.3513513513513513, | |
| "grad_norm": 3.3473320785124807, | |
| "learning_rate": 6.174430228904919e-07, | |
| "loss": 0.8186, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.3513513513513513, | |
| "eval_loss": 0.8826867341995239, | |
| "eval_runtime": 36.3957, | |
| "eval_samples_per_second": 57.699, | |
| "eval_steps_per_second": 0.907, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.4189189189189189, | |
| "grad_norm": 3.708290081383714, | |
| "learning_rate": 5.8114311904423e-07, | |
| "loss": 0.8265, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 1.4864864864864864, | |
| "grad_norm": 3.561238644912943, | |
| "learning_rate": 5.443979476614674e-07, | |
| "loss": 0.8196, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.554054054054054, | |
| "grad_norm": 3.752106899664045, | |
| "learning_rate": 5.074091454568463e-07, | |
| "loss": 0.8243, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 1.6216216216216215, | |
| "grad_norm": 3.524245248482102, | |
| "learning_rate": 4.703796860531429e-07, | |
| "loss": 0.8, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.689189189189189, | |
| "grad_norm": 3.626113759504586, | |
| "learning_rate": 4.3351276617684285e-07, | |
| "loss": 0.8235, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.7567567567567568, | |
| "grad_norm": 3.555163306181175, | |
| "learning_rate": 3.970106906294509e-07, | |
| "loss": 0.8176, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.8243243243243243, | |
| "grad_norm": 3.5936443942910485, | |
| "learning_rate": 3.610737621531781e-07, | |
| "loss": 0.7933, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 1.8918918918918919, | |
| "grad_norm": 3.670717085034342, | |
| "learning_rate": 3.2589918228280066e-07, | |
| "loss": 0.8117, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.9594594594594594, | |
| "grad_norm": 3.4707422128547942, | |
| "learning_rate": 2.916799692151884e-07, | |
| "loss": 0.8011, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 2.027027027027027, | |
| "grad_norm": 3.6322557161041504, | |
| "learning_rate": 2.5860389863462763e-07, | |
| "loss": 0.8014, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.027027027027027, | |
| "eval_loss": 0.8658075332641602, | |
| "eval_runtime": 36.3437, | |
| "eval_samples_per_second": 57.782, | |
| "eval_steps_per_second": 0.908, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.0945945945945947, | |
| "grad_norm": 3.8365084383169954, | |
| "learning_rate": 2.2685247330608414e-07, | |
| "loss": 0.7716, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 2.1621621621621623, | |
| "grad_norm": 3.6658531003049237, | |
| "learning_rate": 1.9659992709070344e-07, | |
| "loss": 0.7524, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.22972972972973, | |
| "grad_norm": 3.687250815077673, | |
| "learning_rate": 1.6801226884893893e-07, | |
| "loss": 0.7735, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 2.2972972972972974, | |
| "grad_norm": 3.5931982468186354, | |
| "learning_rate": 1.412463714778343e-07, | |
| "loss": 0.7535, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.364864864864865, | |
| "grad_norm": 3.726827517069166, | |
| "learning_rate": 1.1644911108130434e-07, | |
| "loss": 0.7456, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 2.4324324324324325, | |
| "grad_norm": 3.6188523000860955, | |
| "learning_rate": 9.375656099715934e-08, | |
| "loss": 0.7546, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 3.5682831997451228, | |
| "learning_rate": 7.329324510360269e-08, | |
| "loss": 0.745, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 2.5675675675675675, | |
| "grad_norm": 3.6369152466203434, | |
| "learning_rate": 5.517145450262639e-08, | |
| "loss": 0.7699, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.635135135135135, | |
| "grad_norm": 3.604157345726647, | |
| "learning_rate": 3.9490631329964554e-08, | |
| "loss": 0.7461, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 2.7027027027027026, | |
| "grad_norm": 3.6723873051705356, | |
| "learning_rate": 2.63368230729043e-08, | |
| "loss": 0.7437, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.7027027027027026, | |
| "eval_loss": 0.8665739297866821, | |
| "eval_runtime": 36.3505, | |
| "eval_samples_per_second": 57.771, | |
| "eval_steps_per_second": 0.908, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.77027027027027, | |
| "grad_norm": 3.421061261868487, | |
| "learning_rate": 1.5782210390350713e-08, | |
| "loss": 0.7692, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 2.8378378378378377, | |
| "grad_norm": 3.6288030361158103, | |
| "learning_rate": 7.884711026201584e-09, | |
| "loss": 0.7482, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.9054054054054053, | |
| "grad_norm": 3.4807112881688616, | |
| "learning_rate": 2.687661989531964e-09, | |
| "loss": 0.7521, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 2.972972972972973, | |
| "grad_norm": 3.773156943868378, | |
| "learning_rate": 2.1958174560282594e-10, | |
| "loss": 0.7664, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 222, | |
| "total_flos": 1308778674782208.0, | |
| "train_loss": 0.8880831481100203, | |
| "train_runtime": 3084.8066, | |
| "train_samples_per_second": 18.38, | |
| "train_steps_per_second": 0.072 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 222, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 10000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1308778674782208.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |