| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9935483870967743, | |
| "eval_steps": 500, | |
| "global_step": 348, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08602150537634409, | |
| "grad_norm": 1.3349374532699585, | |
| "learning_rate": 2.9938918800982563e-05, | |
| "loss": 0.1805, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.17204301075268819, | |
| "grad_norm": 1.0496348142623901, | |
| "learning_rate": 2.975617265898004e-05, | |
| "loss": 0.0827, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.25806451612903225, | |
| "grad_norm": 1.1902973651885986, | |
| "learning_rate": 2.9453249887788343e-05, | |
| "loss": 0.0722, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.34408602150537637, | |
| "grad_norm": 0.8137356638908386, | |
| "learning_rate": 2.9032617538884018e-05, | |
| "loss": 0.0637, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.43010752688172044, | |
| "grad_norm": 0.8646667003631592, | |
| "learning_rate": 2.84977013093626e-05, | |
| "loss": 0.0646, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.5161290322580645, | |
| "grad_norm": 1.2397544384002686, | |
| "learning_rate": 2.7852857642513838e-05, | |
| "loss": 0.0619, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.6021505376344086, | |
| "grad_norm": 0.8160271048545837, | |
| "learning_rate": 2.7103338248251055e-05, | |
| "loss": 0.0588, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.6881720430107527, | |
| "grad_norm": 0.8985416293144226, | |
| "learning_rate": 2.6255247332346036e-05, | |
| "loss": 0.0594, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.7741935483870968, | |
| "grad_norm": 0.968460738658905, | |
| "learning_rate": 2.531549188280135e-05, | |
| "loss": 0.059, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.8602150537634409, | |
| "grad_norm": 0.7749939560890198, | |
| "learning_rate": 2.4291725418235848e-05, | |
| "loss": 0.0557, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.946236559139785, | |
| "grad_norm": 0.6407869458198547, | |
| "learning_rate": 2.3192285656405456e-05, | |
| "loss": 0.0493, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.032258064516129, | |
| "grad_norm": 1.2854279279708862, | |
| "learning_rate": 2.2026126610496852e-05, | |
| "loss": 0.0509, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.118279569892473, | |
| "grad_norm": 0.60258948802948, | |
| "learning_rate": 2.0802745666212592e-05, | |
| "loss": 0.0419, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.2043010752688172, | |
| "grad_norm": 0.9239497780799866, | |
| "learning_rate": 1.953210623354359e-05, | |
| "loss": 0.042, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.2903225806451613, | |
| "grad_norm": 0.7352257966995239, | |
| "learning_rate": 1.8224556603165363e-05, | |
| "loss": 0.0432, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.3763440860215055, | |
| "grad_norm": 0.8709537386894226, | |
| "learning_rate": 1.689074566830434e-05, | |
| "loss": 0.0475, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.4623655913978495, | |
| "grad_norm": 0.7810623049736023, | |
| "learning_rate": 1.5541536198449044e-05, | |
| "loss": 0.0448, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.5483870967741935, | |
| "grad_norm": 0.6514085531234741, | |
| "learning_rate": 1.4187916371218739e-05, | |
| "loss": 0.0407, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.6344086021505375, | |
| "grad_norm": 0.8177109956741333, | |
| "learning_rate": 1.2840910282888211e-05, | |
| "loss": 0.0421, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.7204301075268817, | |
| "grad_norm": 0.9916985630989075, | |
| "learning_rate": 1.1511488166385349e-05, | |
| "loss": 0.0401, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.8064516129032258, | |
| "grad_norm": 1.0323394536972046, | |
| "learning_rate": 1.0210477047960303e-05, | |
| "loss": 0.0438, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.89247311827957, | |
| "grad_norm": 0.920881450176239, | |
| "learning_rate": 8.948472570152874e-06, | |
| "loss": 0.0444, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.978494623655914, | |
| "grad_norm": 1.1360812187194824, | |
| "learning_rate": 7.735752699185711e-06, | |
| "loss": 0.041, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.064516129032258, | |
| "grad_norm": 0.8304557800292969, | |
| "learning_rate": 6.582194019564266e-06, | |
| "loss": 0.0317, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.150537634408602, | |
| "grad_norm": 0.8478759527206421, | |
| "learning_rate": 5.497191297593647e-06, | |
| "loss": 0.0357, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.236559139784946, | |
| "grad_norm": 1.0847315788269043, | |
| "learning_rate": 4.4895809688998655e-06, | |
| "loss": 0.0386, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.3225806451612905, | |
| "grad_norm": 0.6213060617446899, | |
| "learning_rate": 3.567569173085455e-06, | |
| "loss": 0.0289, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.4086021505376345, | |
| "grad_norm": 1.0422612428665161, | |
| "learning_rate": 2.7386649216166233e-06, | |
| "loss": 0.0307, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.4946236559139785, | |
| "grad_norm": 0.9629487991333008, | |
| "learning_rate": 2.0096189432334194e-06, | |
| "loss": 0.0338, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.5806451612903225, | |
| "grad_norm": 0.7275118827819824, | |
| "learning_rate": 1.3863687049356465e-06, | |
| "loss": 0.0279, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.6666666666666665, | |
| "grad_norm": 0.9140329360961914, | |
| "learning_rate": 8.7399005630238e-07, | |
| "loss": 0.0321, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.752688172043011, | |
| "grad_norm": 1.0714858770370483, | |
| "learning_rate": 4.766558909615504e-07, | |
| "loss": 0.0352, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.838709677419355, | |
| "grad_norm": 0.843151330947876, | |
| "learning_rate": 1.9760216187710788e-07, | |
| "loss": 0.0341, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.924731182795699, | |
| "grad_norm": 0.6754664182662964, | |
| "learning_rate": 3.910152723075322e-08, | |
| "loss": 0.0315, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.9935483870967743, | |
| "step": 348, | |
| "total_flos": 2.2441603085736346e+17, | |
| "train_loss": 0.04925085334428426, | |
| "train_runtime": 13305.7708, | |
| "train_samples_per_second": 0.838, | |
| "train_steps_per_second": 0.026 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 348, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 1000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.2441603085736346e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |