| { | |
| "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": 4.215126363518053, | |
| "learning_rate": 1.6666666666666667e-06, | |
| "loss": 0.9296, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.1111111111111111, | |
| "grad_norm": 5.161306302882389, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 0.9489, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.16666666666666666, | |
| "grad_norm": 4.163933865026211, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8764, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.2222222222222222, | |
| "grad_norm": 3.1913354911973144, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.8199, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.2777777777777778, | |
| "grad_norm": 1.993006786586104, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.8011, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.3333333333333333, | |
| "grad_norm": 2.37592942876661, | |
| "learning_rate": 1e-05, | |
| "loss": 0.777, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.3888888888888889, | |
| "grad_norm": 1.97009929150642, | |
| "learning_rate": 9.989294616193018e-06, | |
| "loss": 0.7897, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.4444444444444444, | |
| "grad_norm": 1.904965670883766, | |
| "learning_rate": 9.957224306869053e-06, | |
| "loss": 0.7804, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 1.21854805914713, | |
| "learning_rate": 9.903926402016153e-06, | |
| "loss": 0.7429, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.5555555555555556, | |
| "grad_norm": 1.171424487644955, | |
| "learning_rate": 9.829629131445342e-06, | |
| "loss": 0.8034, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.6111111111111112, | |
| "grad_norm": 1.001721725993888, | |
| "learning_rate": 9.73465064747553e-06, | |
| "loss": 0.7653, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.6666666666666666, | |
| "grad_norm": 3.2983121767179475, | |
| "learning_rate": 9.619397662556434e-06, | |
| "loss": 0.7319, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.7222222222222222, | |
| "grad_norm": 1.0836583095603012, | |
| "learning_rate": 9.484363707663443e-06, | |
| "loss": 0.7718, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.7777777777777778, | |
| "grad_norm": 1.0442034066177648, | |
| "learning_rate": 9.330127018922195e-06, | |
| "loss": 0.6978, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.8333333333333334, | |
| "grad_norm": 1.0264719759153522, | |
| "learning_rate": 9.157348061512728e-06, | |
| "loss": 0.7749, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.8888888888888888, | |
| "grad_norm": 1.0702050666354057, | |
| "learning_rate": 8.966766701456177e-06, | |
| "loss": 0.7806, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.9444444444444444, | |
| "grad_norm": 0.8520315479201978, | |
| "learning_rate": 8.759199037394888e-06, | |
| "loss": 0.7753, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.884331643058831, | |
| "learning_rate": 8.535533905932739e-06, | |
| "loss": 0.7088, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.0555555555555556, | |
| "grad_norm": 0.9511386189970726, | |
| "learning_rate": 8.296729075500345e-06, | |
| "loss": 0.6731, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.1111111111111112, | |
| "grad_norm": 0.9082632017980931, | |
| "learning_rate": 8.043807145043604e-06, | |
| "loss": 0.6305, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.1666666666666667, | |
| "grad_norm": 0.8629684966678431, | |
| "learning_rate": 7.777851165098012e-06, | |
| "loss": 0.664, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.2222222222222223, | |
| "grad_norm": 0.8581737966273371, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.6263, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.2777777777777777, | |
| "grad_norm": 0.8145800427945419, | |
| "learning_rate": 7.211443451095007e-06, | |
| "loss": 0.6148, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.3333333333333333, | |
| "grad_norm": 0.7879794757339875, | |
| "learning_rate": 6.913417161825449e-06, | |
| "loss": 0.6488, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.3888888888888888, | |
| "grad_norm": 0.8058128281377883, | |
| "learning_rate": 6.607197326515808e-06, | |
| "loss": 0.6497, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.4444444444444444, | |
| "grad_norm": 0.7799201229625429, | |
| "learning_rate": 6.294095225512604e-06, | |
| "loss": 0.6491, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 0.7901739399967681, | |
| "learning_rate": 5.975451610080643e-06, | |
| "loss": 0.6274, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.5555555555555556, | |
| "grad_norm": 0.7497271257160177, | |
| "learning_rate": 5.65263096110026e-06, | |
| "loss": 0.6064, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.6111111111111112, | |
| "grad_norm": 0.6938181230390819, | |
| "learning_rate": 5.327015646150716e-06, | |
| "loss": 0.5935, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.6666666666666665, | |
| "grad_norm": 0.7333265469093474, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6162, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.7222222222222223, | |
| "grad_norm": 0.6954845047164014, | |
| "learning_rate": 4.672984353849285e-06, | |
| "loss": 0.5882, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.7777777777777777, | |
| "grad_norm": 0.7083916336241823, | |
| "learning_rate": 4.347369038899744e-06, | |
| "loss": 0.6176, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.8333333333333335, | |
| "grad_norm": 0.72573161274316, | |
| "learning_rate": 4.02454838991936e-06, | |
| "loss": 0.6553, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.8888888888888888, | |
| "grad_norm": 0.7253496499291197, | |
| "learning_rate": 3.705904774487396e-06, | |
| "loss": 0.661, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.9444444444444444, | |
| "grad_norm": 0.6738618246159321, | |
| "learning_rate": 3.3928026734841935e-06, | |
| "loss": 0.6326, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.6810902985229023, | |
| "learning_rate": 3.0865828381745515e-06, | |
| "loss": 0.6303, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 2.0555555555555554, | |
| "grad_norm": 0.6614583167802665, | |
| "learning_rate": 2.7885565489049948e-06, | |
| "loss": 0.5712, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 2.111111111111111, | |
| "grad_norm": 0.679430951773645, | |
| "learning_rate": 2.5000000000000015e-06, | |
| "loss": 0.5289, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 2.1666666666666665, | |
| "grad_norm": 0.6289591102331913, | |
| "learning_rate": 2.2221488349019903e-06, | |
| "loss": 0.538, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 2.2222222222222223, | |
| "grad_norm": 0.657399002632617, | |
| "learning_rate": 1.956192854956397e-06, | |
| "loss": 0.5536, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.2777777777777777, | |
| "grad_norm": 0.7204285646804564, | |
| "learning_rate": 1.7032709244996559e-06, | |
| "loss": 0.5437, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 2.3333333333333335, | |
| "grad_norm": 0.6575697001304656, | |
| "learning_rate": 1.4644660940672628e-06, | |
| "loss": 0.5077, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 2.388888888888889, | |
| "grad_norm": 0.6402892732466081, | |
| "learning_rate": 1.2408009626051137e-06, | |
| "loss": 0.5717, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 2.4444444444444446, | |
| "grad_norm": 0.6363933830535872, | |
| "learning_rate": 1.0332332985438248e-06, | |
| "loss": 0.57, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 0.6472850607083168, | |
| "learning_rate": 8.426519384872733e-07, | |
| "loss": 0.5824, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 2.5555555555555554, | |
| "grad_norm": 0.6316928935934372, | |
| "learning_rate": 6.698729810778065e-07, | |
| "loss": 0.5363, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 2.611111111111111, | |
| "grad_norm": 0.5895456863099023, | |
| "learning_rate": 5.156362923365587e-07, | |
| "loss": 0.5187, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 2.6666666666666665, | |
| "grad_norm": 0.6197672156785694, | |
| "learning_rate": 3.8060233744356634e-07, | |
| "loss": 0.5385, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 2.7222222222222223, | |
| "grad_norm": 0.6079178458993518, | |
| "learning_rate": 2.653493525244721e-07, | |
| "loss": 0.5289, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 2.7777777777777777, | |
| "grad_norm": 0.600339897841452, | |
| "learning_rate": 1.7037086855465902e-07, | |
| "loss": 0.5454, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.8333333333333335, | |
| "grad_norm": 0.594956884732888, | |
| "learning_rate": 9.607359798384785e-08, | |
| "loss": 0.5589, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 2.888888888888889, | |
| "grad_norm": 0.635391896942138, | |
| "learning_rate": 4.2775693130948094e-08, | |
| "loss": 0.5837, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 2.9444444444444446, | |
| "grad_norm": 0.666330714228389, | |
| "learning_rate": 1.0705383806982606e-08, | |
| "loss": 0.5505, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.6422190117878835, | |
| "learning_rate": 0.0, | |
| "loss": 0.5573, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 54, | |
| "total_flos": 2.895189468040397e+16, | |
| "train_loss": 0.6582588244367529, | |
| "train_runtime": 1083.1699, | |
| "train_samples_per_second": 4.767, | |
| "train_steps_per_second": 0.05 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 54, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 10, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.895189468040397e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |