| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 90.0, | |
| "eval_steps": 500, | |
| "global_step": 360, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 0.5093985795974731, | |
| "learning_rate": 0.0009980973490458728, | |
| "loss": 2.5467, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.27049773931503296, | |
| "learning_rate": 0.000992403876506104, | |
| "loss": 1.618, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "grad_norm": 0.22319507598876953, | |
| "learning_rate": 0.0009829629131445341, | |
| "loss": 1.3933, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.2564426362514496, | |
| "learning_rate": 0.0009698463103929542, | |
| "loss": 1.2537, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "grad_norm": 0.29759547114372253, | |
| "learning_rate": 0.0009531538935183251, | |
| "loss": 1.1368, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.39310407638549805, | |
| "learning_rate": 0.0009330127018922195, | |
| "loss": 1.0137, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 17.5, | |
| "grad_norm": 0.4282013475894928, | |
| "learning_rate": 0.0009095760221444959, | |
| "loss": 0.9095, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 0.5433480143547058, | |
| "learning_rate": 0.000883022221559489, | |
| "loss": 0.8259, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 22.5, | |
| "grad_norm": 0.5130556225776672, | |
| "learning_rate": 0.0008535533905932737, | |
| "loss": 0.7432, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "grad_norm": 0.5033867359161377, | |
| "learning_rate": 0.0008213938048432696, | |
| "loss": 0.6756, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 27.5, | |
| "grad_norm": 0.5322891473770142, | |
| "learning_rate": 0.0007867882181755231, | |
| "loss": 0.6123, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 0.5658751726150513, | |
| "learning_rate": 0.00075, | |
| "loss": 0.567, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 32.5, | |
| "grad_norm": 0.5688820481300354, | |
| "learning_rate": 0.0007113091308703497, | |
| "loss": 0.5171, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "grad_norm": 0.48720425367355347, | |
| "learning_rate": 0.0006710100716628344, | |
| "loss": 0.4825, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 37.5, | |
| "grad_norm": 0.5325213074684143, | |
| "learning_rate": 0.0006294095225512603, | |
| "loss": 0.4498, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 0.527807891368866, | |
| "learning_rate": 0.0005868240888334653, | |
| "loss": 0.4184, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 42.5, | |
| "grad_norm": 0.4882418215274811, | |
| "learning_rate": 0.0005435778713738292, | |
| "loss": 0.3946, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "grad_norm": 0.49641069769859314, | |
| "learning_rate": 0.0005, | |
| "loss": 0.3706, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 47.5, | |
| "grad_norm": 0.5216576457023621, | |
| "learning_rate": 0.00045642212862617086, | |
| "loss": 0.3529, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "grad_norm": 0.5289739966392517, | |
| "learning_rate": 0.00041317591116653486, | |
| "loss": 0.3334, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 52.5, | |
| "grad_norm": 0.49065396189689636, | |
| "learning_rate": 0.0003705904774487396, | |
| "loss": 0.3166, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "grad_norm": 0.4922383725643158, | |
| "learning_rate": 0.0003289899283371657, | |
| "loss": 0.3086, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 57.5, | |
| "grad_norm": 0.42886731028556824, | |
| "learning_rate": 0.0002886908691296504, | |
| "loss": 0.2894, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "grad_norm": 0.42780551314353943, | |
| "learning_rate": 0.0002500000000000001, | |
| "loss": 0.286, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 62.5, | |
| "grad_norm": 0.4010085463523865, | |
| "learning_rate": 0.00021321178182447708, | |
| "loss": 0.2775, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "grad_norm": 0.4402850866317749, | |
| "learning_rate": 0.0001786061951567303, | |
| "loss": 0.2714, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 67.5, | |
| "grad_norm": 0.4067270755767822, | |
| "learning_rate": 0.00014644660940672628, | |
| "loss": 0.2664, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "grad_norm": 0.43572983145713806, | |
| "learning_rate": 0.00011697777844051105, | |
| "loss": 0.2582, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 72.5, | |
| "grad_norm": 0.3881024420261383, | |
| "learning_rate": 9.042397785550405e-05, | |
| "loss": 0.2549, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "grad_norm": 0.3768531382083893, | |
| "learning_rate": 6.698729810778065e-05, | |
| "loss": 0.251, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 77.5, | |
| "grad_norm": 0.3921433389186859, | |
| "learning_rate": 4.684610648167503e-05, | |
| "loss": 0.2509, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "grad_norm": 0.38551065325737, | |
| "learning_rate": 3.0153689607045842e-05, | |
| "loss": 0.2466, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 82.5, | |
| "grad_norm": 0.3801429867744446, | |
| "learning_rate": 1.70370868554659e-05, | |
| "loss": 0.2462, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 85.0, | |
| "grad_norm": 0.39348262548446655, | |
| "learning_rate": 7.59612349389599e-06, | |
| "loss": 0.2468, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 87.5, | |
| "grad_norm": 0.3822017014026642, | |
| "learning_rate": 1.9026509541272275e-06, | |
| "loss": 0.2477, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "grad_norm": 0.3794040381908417, | |
| "learning_rate": 0.0, | |
| "loss": 0.2449, | |
| "step": 360 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 360, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 90, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.43272328822784e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |