| { | |
| "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": 6.17005626198819, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 1.0472, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 6.438107322073817, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 1.0688, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 6.071434295023916, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 1.0774, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 2.707764630640578, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.9552, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 3.711543915769142, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.9503, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 4.079489426888842, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.9622, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 4.107721935177619, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9282, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 2.9805074251923473, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.8884, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 2.335927749227063, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.8651, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 1.7147941401016604, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.8352, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 2.102581794428384, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 1.2739, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 1.259159154545842, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.758, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 1.1824834086966654, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.784, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 0.9486932948585013, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.7424, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 1.131351213208882, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.7964, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 0.8323749991080968, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.7107, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 1.00619470517082, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.7365, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 0.8353205127670753, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.6698, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 0.9105767690662245, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.7341, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 1.0336010824718762, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.6919, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 1.469002037372581, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 1.1327, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 0.8079151558221264, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.5794, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 0.7997758071616262, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.619, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 0.8676432070049109, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.6241, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.8013138080335881, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.6198, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 0.6323582610620165, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.5638, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.6974892013081477, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.5805, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.6062490593383558, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.5924, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.6702893898422643, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.6092, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.6463004339502111, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.5858, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 0.7074215792010169, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.7238, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 0.8326965044738764, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.7453, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 0.6031771501067653, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.5567, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.2640000000000002, | |
| "grad_norm": 0.5621113253379892, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.486, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.6059187623833876, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.5218, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.456, | |
| "grad_norm": 0.5643621805237538, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.4751, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.552, | |
| "grad_norm": 0.5991214055298102, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.5196, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.648, | |
| "grad_norm": 0.5504277661729509, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.4755, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7439999999999998, | |
| "grad_norm": 0.5195620693940317, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.4675, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.5567554635187709, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.4732, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.936, | |
| "grad_norm": 0.6371664368841761, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.5144, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.032, | |
| "grad_norm": 0.8457067013368751, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.7682, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 0.48383057925399775, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.3986, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.224, | |
| "grad_norm": 0.6454519851764626, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.493, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.46101244556202525, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.3987, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.416, | |
| "grad_norm": 0.5379713842192534, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.3921, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.5120000000000005, | |
| "grad_norm": 0.6335155392784169, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.4129, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 0.5745539978526327, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.4032, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.704, | |
| "grad_norm": 0.49788511433793875, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.4363, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.4373147932095772, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.3733, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.896, | |
| "grad_norm": 0.4818494055366588, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.4347, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.992, | |
| "grad_norm": 0.7366124367841175, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.6164, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.088, | |
| "grad_norm": 0.5289520484815684, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.4127, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.184, | |
| "grad_norm": 0.4925941075892466, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.3602, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.490264607812914, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.3673, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.376, | |
| "grad_norm": 0.4603303165601469, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.3661, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.4719999999999995, | |
| "grad_norm": 0.4489011601087028, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.3499, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.568, | |
| "grad_norm": 0.45306820235621026, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.3933, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.664, | |
| "grad_norm": 0.42064579982611183, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.3623, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.44572909857551013, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.3784, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.856, | |
| "grad_norm": 0.41815357068390685, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.379, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.952, | |
| "grad_norm": 0.41098618661493447, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.3637, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.048, | |
| "grad_norm": 0.69977280916362, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.5768, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.144, | |
| "grad_norm": 0.38070543444988303, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.3495, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.3899069533619493, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.3812, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.336, | |
| "grad_norm": 0.3414734728106033, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.3171, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.432, | |
| "grad_norm": 0.3743232407915419, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.3698, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5280000000000005, | |
| "grad_norm": 0.37041780678658287, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.3487, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.624, | |
| "grad_norm": 0.39208305889306666, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.3879, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.35583565023783154, | |
| "learning_rate": 0.0, | |
| "loss": 0.3121, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 70, | |
| "total_flos": 1.2045193053418291e+17, | |
| "train_loss": 0.5977857010705131, | |
| "train_runtime": 5167.7074, | |
| "train_samples_per_second": 1.355, | |
| "train_steps_per_second": 0.014 | |
| } | |
| ], | |
| "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": 1.2045193053418291e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |