| { | |
| "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": 6.604844489673478, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 1.0498, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 6.7832198408878215, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 1.0566, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 6.11888948000492, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 1.0354, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 3.8771235601682514, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 1.0411, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 4.0501641658881615, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.9616, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 4.051429236877689, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.9529, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 4.139228954764867, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9314, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 3.0329502437257903, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.8996, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 2.337673016414626, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.8592, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 2.0419888128736248, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.8411, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 1.6099764383285402, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 0.8139, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 1.1659999086628976, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.7805, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 1.325141274620268, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.7637, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 2.5453668607970483, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.7329, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 1.2821253853248127, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.7515, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 0.878416680079174, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.7429, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 1.25506102734324, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.7352, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 0.9426702671160289, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.7196, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 0.9999170113082805, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.7095, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.818575561180334, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.713, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 0.6854539165930189, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.6956, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 0.8158405000908213, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.6748, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 0.682132282971258, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.6648, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 0.6624708858712745, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.6548, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.5761072414893121, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.6792, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 6.6840341021834995, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.6968, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.9215964035298393, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.6503, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.5945610095182325, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.6455, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.8973075144566792, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.6506, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.6099785873149247, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.637, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 0.8082178028570961, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.6301, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 0.7516735297770826, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.6197, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 0.5155897516771877, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.6053, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.2640000000000002, | |
| "grad_norm": 0.6210730047544852, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.6101, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.5274240292051746, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.6064, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.456, | |
| "grad_norm": 0.4455008341979113, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.5853, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.552, | |
| "grad_norm": 0.5775468137636318, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.5983, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.648, | |
| "grad_norm": 0.4586232174588804, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.5872, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7439999999999998, | |
| "grad_norm": 0.43977520366953, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.5728, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.5306704101838037, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.5944, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.936, | |
| "grad_norm": 0.49246541485485185, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.5829, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.032, | |
| "grad_norm": 0.43129176733170826, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.5896, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 0.4056065811473056, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.566, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.224, | |
| "grad_norm": 0.4481733652966385, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.5505, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.37794562535448456, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.5462, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.416, | |
| "grad_norm": 0.39389036589608717, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.5732, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.5120000000000005, | |
| "grad_norm": 0.3483344540798747, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.5436, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 0.3909169730559128, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.5416, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.704, | |
| "grad_norm": 0.37452946176136903, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.5561, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.3425714701131914, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.5626, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.896, | |
| "grad_norm": 0.35632724558474527, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.5426, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.992, | |
| "grad_norm": 0.3657549546734331, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.5493, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.088, | |
| "grad_norm": 0.3201546566268174, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.5429, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.184, | |
| "grad_norm": 0.31269621075272863, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.5462, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.32911958215631437, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.5179, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.376, | |
| "grad_norm": 0.3259417021765892, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.5296, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.4719999999999995, | |
| "grad_norm": 0.30489159733950405, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.5381, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.568, | |
| "grad_norm": 0.32531099447599704, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.5507, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.664, | |
| "grad_norm": 0.32043473075073575, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.539, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.3049083270415507, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.5313, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.856, | |
| "grad_norm": 0.2724933793793105, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.5175, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.952, | |
| "grad_norm": 0.2730824013192508, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.5251, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.048, | |
| "grad_norm": 0.3007013484732537, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.5123, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.144, | |
| "grad_norm": 0.2562568125151116, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.5244, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.27145084752693266, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.525, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.336, | |
| "grad_norm": 0.2610024344297741, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.5291, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.432, | |
| "grad_norm": 0.2650714297856863, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.4929, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5280000000000005, | |
| "grad_norm": 0.26223592775817706, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.5308, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.624, | |
| "grad_norm": 0.2639700489271648, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.5203, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.27007886224853306, | |
| "learning_rate": 0.0, | |
| "loss": 0.5348, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 70, | |
| "total_flos": 2.5376204695116186e+17, | |
| "train_loss": 0.6566088931901114, | |
| "train_runtime": 8555.0108, | |
| "train_samples_per_second": 0.818, | |
| "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": 2.5376204695116186e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |