| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 54, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05555555555555555, | |
| "grad_norm": 6.603050260549088, | |
| "learning_rate": 1.6666666666666667e-06, | |
| "loss": 0.5685, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.1111111111111111, | |
| "grad_norm": 5.445146526347019, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 0.495, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.16666666666666666, | |
| "grad_norm": 5.609367418806909, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5354, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.2222222222222222, | |
| "grad_norm": 4.306546116247042, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.5173, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.2777777777777778, | |
| "grad_norm": 3.4262833071668677, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.4717, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.3333333333333333, | |
| "grad_norm": 3.9286566875492754, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5151, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.3888888888888889, | |
| "grad_norm": 5.489531264932943, | |
| "learning_rate": 9.989294616193018e-06, | |
| "loss": 0.494, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.4444444444444444, | |
| "grad_norm": 4.282143799678189, | |
| "learning_rate": 9.957224306869053e-06, | |
| "loss": 0.4532, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 3.830449207187074, | |
| "learning_rate": 9.903926402016153e-06, | |
| "loss": 0.4669, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.5555555555555556, | |
| "grad_norm": 3.4283032271217246, | |
| "learning_rate": 9.829629131445342e-06, | |
| "loss": 0.4627, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.6111111111111112, | |
| "grad_norm": 2.910042689612452, | |
| "learning_rate": 9.73465064747553e-06, | |
| "loss": 0.4031, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.6666666666666666, | |
| "grad_norm": 3.19592750447562, | |
| "learning_rate": 9.619397662556434e-06, | |
| "loss": 0.4516, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.7222222222222222, | |
| "grad_norm": 2.567761028710531, | |
| "learning_rate": 9.484363707663443e-06, | |
| "loss": 0.3436, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.7777777777777778, | |
| "grad_norm": 2.933256204770112, | |
| "learning_rate": 9.330127018922195e-06, | |
| "loss": 0.3995, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.8333333333333334, | |
| "grad_norm": 2.3849458808311264, | |
| "learning_rate": 9.157348061512728e-06, | |
| "loss": 0.2984, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.8888888888888888, | |
| "grad_norm": 3.1697604204843444, | |
| "learning_rate": 8.966766701456177e-06, | |
| "loss": 0.4526, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.9444444444444444, | |
| "grad_norm": 2.6410409017659795, | |
| "learning_rate": 8.759199037394888e-06, | |
| "loss": 0.4155, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 2.6860078740348237, | |
| "learning_rate": 8.535533905932739e-06, | |
| "loss": 0.3603, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.0555555555555556, | |
| "grad_norm": 2.3523753662302904, | |
| "learning_rate": 8.296729075500345e-06, | |
| "loss": 0.2709, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.1111111111111112, | |
| "grad_norm": 2.1297592085294514, | |
| "learning_rate": 8.043807145043604e-06, | |
| "loss": 0.2282, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.1666666666666667, | |
| "grad_norm": 1.8053386362884114, | |
| "learning_rate": 7.777851165098012e-06, | |
| "loss": 0.2134, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.2222222222222223, | |
| "grad_norm": 1.9862232463835396, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.1635, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.2777777777777777, | |
| "grad_norm": 2.160681684352547, | |
| "learning_rate": 7.211443451095007e-06, | |
| "loss": 0.2416, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.3333333333333333, | |
| "grad_norm": 2.46016642561212, | |
| "learning_rate": 6.913417161825449e-06, | |
| "loss": 0.2667, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.3888888888888888, | |
| "grad_norm": 2.167009937420834, | |
| "learning_rate": 6.607197326515808e-06, | |
| "loss": 0.1659, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.4444444444444444, | |
| "grad_norm": 2.509921799553618, | |
| "learning_rate": 6.294095225512604e-06, | |
| "loss": 0.1686, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 2.570429240697162, | |
| "learning_rate": 5.975451610080643e-06, | |
| "loss": 0.2028, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.5555555555555556, | |
| "grad_norm": 2.1747066459504247, | |
| "learning_rate": 5.65263096110026e-06, | |
| "loss": 0.1826, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.6111111111111112, | |
| "grad_norm": 2.5209483729196154, | |
| "learning_rate": 5.327015646150716e-06, | |
| "loss": 0.2438, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.6666666666666665, | |
| "grad_norm": 2.1584467581976496, | |
| "learning_rate": 5e-06, | |
| "loss": 0.21, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.7222222222222223, | |
| "grad_norm": 2.1517545812849828, | |
| "learning_rate": 4.672984353849285e-06, | |
| "loss": 0.1968, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.7777777777777777, | |
| "grad_norm": 2.372471554131305, | |
| "learning_rate": 4.347369038899744e-06, | |
| "loss": 0.2078, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.8333333333333335, | |
| "grad_norm": 2.2276822325726546, | |
| "learning_rate": 4.02454838991936e-06, | |
| "loss": 0.2334, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.8888888888888888, | |
| "grad_norm": 2.203685201438035, | |
| "learning_rate": 3.705904774487396e-06, | |
| "loss": 0.232, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.9444444444444444, | |
| "grad_norm": 2.1810116193964464, | |
| "learning_rate": 3.3928026734841935e-06, | |
| "loss": 0.207, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.5386033372246861, | |
| "learning_rate": 3.0865828381745515e-06, | |
| "loss": 0.1448, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 2.0555555555555554, | |
| "grad_norm": 1.884821380741252, | |
| "learning_rate": 2.7885565489049948e-06, | |
| "loss": 0.1191, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 2.111111111111111, | |
| "grad_norm": 1.3470455257212837, | |
| "learning_rate": 2.5000000000000015e-06, | |
| "loss": 0.0879, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 2.1666666666666665, | |
| "grad_norm": 1.6870939478443838, | |
| "learning_rate": 2.2221488349019903e-06, | |
| "loss": 0.109, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 2.2222222222222223, | |
| "grad_norm": 1.543095509192476, | |
| "learning_rate": 1.956192854956397e-06, | |
| "loss": 0.0959, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.2777777777777777, | |
| "grad_norm": 1.4355310847922262, | |
| "learning_rate": 1.7032709244996559e-06, | |
| "loss": 0.0866, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 2.3333333333333335, | |
| "grad_norm": 1.2217586959435303, | |
| "learning_rate": 1.4644660940672628e-06, | |
| "loss": 0.0829, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 2.388888888888889, | |
| "grad_norm": 1.5423699667950068, | |
| "learning_rate": 1.2408009626051137e-06, | |
| "loss": 0.0936, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 2.4444444444444446, | |
| "grad_norm": 1.2238363086013229, | |
| "learning_rate": 1.0332332985438248e-06, | |
| "loss": 0.0846, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 1.4028302679140323, | |
| "learning_rate": 8.426519384872733e-07, | |
| "loss": 0.0923, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 2.5555555555555554, | |
| "grad_norm": 1.6101000022511165, | |
| "learning_rate": 6.698729810778065e-07, | |
| "loss": 0.1137, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 2.611111111111111, | |
| "grad_norm": 1.604022330064103, | |
| "learning_rate": 5.156362923365587e-07, | |
| "loss": 0.087, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 2.6666666666666665, | |
| "grad_norm": 1.2647005113132082, | |
| "learning_rate": 3.8060233744356634e-07, | |
| "loss": 0.0903, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 2.7222222222222223, | |
| "grad_norm": 1.2734035263866237, | |
| "learning_rate": 2.653493525244721e-07, | |
| "loss": 0.088, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 2.7777777777777777, | |
| "grad_norm": 1.4363872051326905, | |
| "learning_rate": 1.7037086855465902e-07, | |
| "loss": 0.0917, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.8333333333333335, | |
| "grad_norm": 1.3002982053513228, | |
| "learning_rate": 9.607359798384785e-08, | |
| "loss": 0.0826, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 2.888888888888889, | |
| "grad_norm": 1.3872970401940254, | |
| "learning_rate": 4.2775693130948094e-08, | |
| "loss": 0.0727, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 2.9444444444444446, | |
| "grad_norm": 1.3308535268449344, | |
| "learning_rate": 1.0705383806982606e-08, | |
| "loss": 0.0702, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.2554651600866564, | |
| "learning_rate": 0.0, | |
| "loss": 0.0724, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 54, | |
| "total_flos": 2100901969920.0, | |
| "train_loss": 0.2500925399363041, | |
| "train_runtime": 303.5482, | |
| "train_samples_per_second": 2.797, | |
| "train_steps_per_second": 0.178 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 54, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "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": 2100901969920.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |