| { | |
| "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": 7.045953147245434, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 1.1635, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 7.140474278562816, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 1.2223, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 6.433055472273257, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 1.144, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 2.753962750623152, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 1.0767, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 4.156093097750324, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 1.0104, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 4.619751917524022, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 1.0191, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 4.866866190373246, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9595, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 4.042771077312268, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.9802, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 3.3251008103663673, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 1.02, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 2.700896052565273, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.8661, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 3.765102329167205, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 1.4141, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 2.2877512781210805, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.8207, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 2.011862448150546, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.7758, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 1.5025912431630202, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.7777, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 1.3335599113338317, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.8096, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 1.4719480957326998, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.7235, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 1.4106698903445756, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.7292, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 1.0146908198833848, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.7144, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 1.2807193696417971, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.8303, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 1.0680990871699445, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.7616, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 1.6828972178935873, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 1.0921, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 0.9051585896161396, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.6724, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 1.0095763594566753, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.5657, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 0.9743770196120521, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.5474, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 1.2016075375158415, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.6473, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 1.1146710483098126, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.5168, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.8828747302294339, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.5159, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.8417373418871406, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.5653, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.7860522512941126, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.5228, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.7647944383306859, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.5105, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 1.0230864737955494, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.7182, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 0.9386694930898264, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.5376, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 0.7520093308603011, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.3854, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.2640000000000002, | |
| "grad_norm": 0.7558884917547274, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.3967, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.873660575459229, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.385, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.456, | |
| "grad_norm": 0.8259884136609715, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.3876, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.552, | |
| "grad_norm": 0.8334373495106804, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.3933, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.648, | |
| "grad_norm": 0.6738344589738631, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.3581, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7439999999999998, | |
| "grad_norm": 0.6480472208768966, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.3435, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.7928514883036736, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.3689, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.936, | |
| "grad_norm": 0.6493434601538668, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.3927, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.032, | |
| "grad_norm": 1.1580640230753043, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.5739, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 0.6042465101723894, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.245, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.224, | |
| "grad_norm": 0.6266224837071126, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.2472, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.6433763543815567, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.2873, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.416, | |
| "grad_norm": 0.6447091706433478, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.2608, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.5120000000000005, | |
| "grad_norm": 0.8233980489048546, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.3055, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 0.6152122352865771, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.2228, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.704, | |
| "grad_norm": 0.5355989260223736, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.216, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.6031688742794519, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.2473, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.896, | |
| "grad_norm": 0.5852353412883003, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.2308, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.992, | |
| "grad_norm": 0.9094752393966428, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.3873, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.088, | |
| "grad_norm": 0.7926652378466514, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.2359, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.184, | |
| "grad_norm": 0.6726865678424423, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.1873, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.5158954505709227, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.197, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.376, | |
| "grad_norm": 0.49880886924316453, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.1894, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.4719999999999995, | |
| "grad_norm": 0.5363977621149951, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.1657, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.568, | |
| "grad_norm": 0.5182932981345756, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.1819, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.664, | |
| "grad_norm": 0.5788686788328858, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.1946, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.6282620228201624, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.1981, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.856, | |
| "grad_norm": 0.5855873918448421, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.1944, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.952, | |
| "grad_norm": 0.5050515733330136, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.1885, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.048, | |
| "grad_norm": 0.8119160386262705, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.2587, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.144, | |
| "grad_norm": 0.408058193161249, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.1607, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.45527849303323825, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.2082, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.336, | |
| "grad_norm": 0.36788899810992925, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.1309, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.432, | |
| "grad_norm": 0.4676299259223216, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.1925, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5280000000000005, | |
| "grad_norm": 0.41579437232924926, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.1474, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.624, | |
| "grad_norm": 0.45360026373649986, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.1755, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.4426261965132341, | |
| "learning_rate": 0.0, | |
| "loss": 0.1482, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 70, | |
| "total_flos": 5.715181934791885e+16, | |
| "train_loss": 0.5174432337284088, | |
| "train_runtime": 2652.1921, | |
| "train_samples_per_second": 2.639, | |
| "train_steps_per_second": 0.026 | |
| } | |
| ], | |
| "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": 5.715181934791885e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |