| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.72, | |
| "eval_steps": 500, | |
| "global_step": 70, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.096, | |
| "grad_norm": 6.279618465887646, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 0.8993, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 6.49571485745033, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 0.8886, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 6.124460821535195, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 0.8862, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 2.653622466398279, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.8138, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 4.206579484291407, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.7982, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 4.6264123133405235, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.7988, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 4.594055901644945, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7706, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 3.441080330094095, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.7635, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 2.835315404043152, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.7209, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 2.579820079512453, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.7388, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 2.717785225669513, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 1.1227, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 1.5547322260744376, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.6719, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 1.3393567930576382, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.6036, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 0.8789867833613715, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.6006, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 0.9101752370128238, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.6078, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 0.8484812691906146, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.5972, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 0.6521689649712946, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.5579, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 1.0569224591541706, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.7111, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 0.6143179366713816, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.5397, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.764087770025139, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.5885, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 1.3396387079071428, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.9332, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 0.5952728518888278, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.4869, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 0.8118126882792409, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.5193, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 0.7133498221368507, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.552, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.6439364032652635, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.4632, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 0.6560856711164174, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.5117, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.6521852975637941, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.4499, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.7092717136333666, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.4981, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.5643586369430377, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.4714, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.5699395019320763, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.5064, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 0.7473684876181278, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.5965, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 0.7431192603768191, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.615, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 0.521561685539083, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.405, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.2640000000000002, | |
| "grad_norm": 0.5860051101542451, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.4052, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.6216106957344598, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.3972, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.456, | |
| "grad_norm": 0.5851007018660533, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.3749, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.552, | |
| "grad_norm": 0.635804666139263, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.4246, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.648, | |
| "grad_norm": 0.6297244416068373, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.4393, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7439999999999998, | |
| "grad_norm": 0.4749125413665247, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.3595, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.5674328227309864, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.3715, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.936, | |
| "grad_norm": 0.6732382184862686, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.3847, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.032, | |
| "grad_norm": 0.8084544602751266, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.6246, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 0.4357842439853899, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.303, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.224, | |
| "grad_norm": 0.6575972716809633, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.3752, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.47239818453077836, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.3215, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.416, | |
| "grad_norm": 0.4355724229940792, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.2853, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.5120000000000005, | |
| "grad_norm": 0.61682930116866, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.363, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 0.43461330636375684, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.2892, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.704, | |
| "grad_norm": 0.48821509458575385, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.3583, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.3979876622655832, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.2677, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.896, | |
| "grad_norm": 0.4763394930846895, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.3653, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.992, | |
| "grad_norm": 0.6334006739456187, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.4537, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.088, | |
| "grad_norm": 0.5266095596206775, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.3278, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.184, | |
| "grad_norm": 0.44128332060974723, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.2706, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.4292728419199483, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.3067, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.376, | |
| "grad_norm": 0.3667588334993992, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.2608, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.4719999999999995, | |
| "grad_norm": 0.4408427477024568, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.2953, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.568, | |
| "grad_norm": 0.37555976031572375, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.2691, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.664, | |
| "grad_norm": 0.44556771608874246, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.301, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.35406698741089837, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.2616, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.856, | |
| "grad_norm": 0.37380979117099244, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.2896, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.952, | |
| "grad_norm": 0.39300189289521714, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.2959, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.048, | |
| "grad_norm": 0.6179960233732275, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.404, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.144, | |
| "grad_norm": 0.35268086202149196, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.2751, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.36486843071065356, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.2666, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.336, | |
| "grad_norm": 0.32121104801257955, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.2541, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.432, | |
| "grad_norm": 0.32137609630583747, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.2793, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5280000000000005, | |
| "grad_norm": 0.35149008390774134, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.275, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.624, | |
| "grad_norm": 0.334592276344663, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.2818, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.34601787597700795, | |
| "learning_rate": 0.0, | |
| "loss": 0.286, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 70, | |
| "total_flos": 1.1913104726582886e+17, | |
| "train_loss": 0.4864661923476628, | |
| "train_runtime": 5735.7432, | |
| "train_samples_per_second": 1.22, | |
| "train_steps_per_second": 0.012 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 70, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "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": 1.1913104726582886e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |