| { | |
| "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": 6.475348885078809, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 1.031, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 6.470206686346939, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 1.0289, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 5.8641649968265765, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 1.0262, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 2.8362263222584443, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.9186, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 4.0934966905579175, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.9415, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 4.197293701621639, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.9563, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 3.840768601516746, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9031, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 2.718866233819792, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.8591, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 2.332692958904956, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.8479, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 1.7422237178797706, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.8147, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 1.9578956523325644, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 1.2602, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 1.7383213696541158, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.7744, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 1.2005780462398976, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.736, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 0.8940872217059948, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.7363, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 1.394357783090997, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.7597, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 0.9774055806962896, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.74, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 1.039895870584419, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.6943, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 1.0226176188243807, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.7361, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 0.9281690655096617, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.7036, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.6408884829979477, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.6805, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 1.2962704611681337, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 1.1026, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 3.0632504838818706, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.6603, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 4.03600022462183, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.6633, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 1.7143218489972314, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.6568, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 2.7501895529808214, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.6531, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 1.3584874795785624, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.6551, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.7368820419147, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.6211, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 1.1751215189646713, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.6243, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.8915938564937985, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.6426, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.7910839718197438, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.6199, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 0.7712472602511662, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.7405, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 1.0684629107152404, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.8524, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 0.5785986592952662, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.6122, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.2640000000000002, | |
| "grad_norm": 0.5551855772477096, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.5727, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.631758721863412, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.5964, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.456, | |
| "grad_norm": 0.48386130959063545, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.5852, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.552, | |
| "grad_norm": 0.49415305665217113, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.5756, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.648, | |
| "grad_norm": 0.4431717690321708, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.5778, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7439999999999998, | |
| "grad_norm": 0.495058061553528, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.585, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.4157130791803711, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.5837, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.936, | |
| "grad_norm": 0.4668661343429709, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.5767, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.032, | |
| "grad_norm": 0.6519783223818135, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.9189, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 0.4065791272570537, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.5391, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.224, | |
| "grad_norm": 0.45749612583514987, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.5334, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.42061670634873033, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.5561, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.416, | |
| "grad_norm": 0.4164937562135911, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.5293, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.5120000000000005, | |
| "grad_norm": 0.44532378456176813, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.5734, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 0.44540593759244573, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.5497, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.704, | |
| "grad_norm": 0.38727159842798814, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.55, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.34813547809638773, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.5151, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.896, | |
| "grad_norm": 0.3624180198348498, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.5208, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.992, | |
| "grad_norm": 0.5746229573424801, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.7725, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.088, | |
| "grad_norm": 0.4273710986717972, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.5916, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.184, | |
| "grad_norm": 0.34234779802517046, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.504, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.37196936781993695, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.5108, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.376, | |
| "grad_norm": 0.36071005295388076, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.5135, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.4719999999999995, | |
| "grad_norm": 0.30594871676332996, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.5162, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.568, | |
| "grad_norm": 0.2928372855256993, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.5085, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.664, | |
| "grad_norm": 0.2848945732783456, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.5122, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.2994624662646981, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.5193, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.856, | |
| "grad_norm": 0.2932830635526158, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.5133, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.952, | |
| "grad_norm": 0.2582572958061305, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.4971, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.048, | |
| "grad_norm": 0.503897494930926, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.8399, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.144, | |
| "grad_norm": 0.27078713673336724, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.5132, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.24934133841499365, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.495, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.336, | |
| "grad_norm": 0.24907219257809815, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.505, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.432, | |
| "grad_norm": 0.2609848132026422, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.4939, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5280000000000005, | |
| "grad_norm": 0.24944272834960846, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.5046, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.624, | |
| "grad_norm": 0.2510610307076265, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.5283, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.24759429705364214, | |
| "learning_rate": 0.0, | |
| "loss": 0.5132, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 70, | |
| "total_flos": 2.5532015569167974e+17, | |
| "train_loss": 0.6720493512494223, | |
| "train_runtime": 9099.4162, | |
| "train_samples_per_second": 0.769, | |
| "train_steps_per_second": 0.008 | |
| } | |
| ], | |
| "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": 2.5532015569167974e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |