| { | |
| "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": 6.673135595971492, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 0.8155, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 7.047701901514139, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 0.8298, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 6.555254425585188, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 0.8198, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 3.086205755770442, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.6921, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 3.9870534201592536, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.6764, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 4.032742067287217, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.6736, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 4.823234709618303, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6185, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 3.7496559340222633, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.6112, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 2.8383797356239207, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.5765, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 2.5126983514246333, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.5501, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 1.4987191779927913, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 0.5217, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 1.5245438571569268, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.5067, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 1.2749409517281813, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.4686, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 0.9214748652017067, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.4775, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 0.9658738622156932, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.4518, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 0.9253143868835119, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.4745, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 0.7092936710111104, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.4635, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 0.7092135382277782, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.4357, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 0.7195318105083511, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.4394, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.6502464889162165, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.46, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 0.6995091868070391, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.4408, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 0.6482173889210819, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.3624, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 0.6345587611230575, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.3643, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 0.6442871469393701, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.3603, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.6611673078300775, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.3358, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 0.6120443973770684, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.3537, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.6449123401192141, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.3473, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.6612228584414032, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.3404, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.6452540872136787, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.3645, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.5425302232549788, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.3279, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 0.7158901121703884, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.3245, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 0.5407849234576103, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.3059, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 0.5731750940129817, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.2872, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.2640000000000002, | |
| "grad_norm": 0.5624826172631466, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.266, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.5052177519741691, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.2457, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.456, | |
| "grad_norm": 0.4883825431478865, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.2426, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.552, | |
| "grad_norm": 0.5337837594020157, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.2957, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.648, | |
| "grad_norm": 0.48907739526587374, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.2473, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7439999999999998, | |
| "grad_norm": 0.43961680807509107, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.2702, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.4744034258922979, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.2237, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.936, | |
| "grad_norm": 0.49063246765903773, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.2563, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.032, | |
| "grad_norm": 0.4707507198094458, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.2312, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 0.4583127746592598, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.2555, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.224, | |
| "grad_norm": 0.46318360388887425, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.2122, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.40140599701254176, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.2007, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.416, | |
| "grad_norm": 0.48991482945850273, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.2322, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.5120000000000005, | |
| "grad_norm": 0.43477941214846444, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.188, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 0.40397025654722485, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.2053, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.704, | |
| "grad_norm": 0.4337675888874392, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.1797, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.38233035541461996, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.1965, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.896, | |
| "grad_norm": 0.3770102554804616, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.1812, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.992, | |
| "grad_norm": 0.4025506327499795, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.2158, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.088, | |
| "grad_norm": 0.3454770618363356, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.1557, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.184, | |
| "grad_norm": 0.3178145618160784, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.1879, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.3399800129529387, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.1747, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.376, | |
| "grad_norm": 0.32347600275491545, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.1873, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.4719999999999995, | |
| "grad_norm": 0.2869810950939557, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.1833, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.568, | |
| "grad_norm": 0.29981479583384235, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.1727, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.664, | |
| "grad_norm": 0.32206783760497215, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.1967, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.3204916295297156, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.1823, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.856, | |
| "grad_norm": 0.3304539497153748, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.1877, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.952, | |
| "grad_norm": 0.29289847893618015, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.1469, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.048, | |
| "grad_norm": 0.27540852057160964, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.1573, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.144, | |
| "grad_norm": 0.2784947348759955, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.1579, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.2769261623965802, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.1788, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.336, | |
| "grad_norm": 0.27738904396972824, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.1761, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.432, | |
| "grad_norm": 0.25821042242945735, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.1666, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5280000000000005, | |
| "grad_norm": 0.28216501957748413, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.1787, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.624, | |
| "grad_norm": 0.27242091577160826, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.165, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.28144047333167166, | |
| "learning_rate": 0.0, | |
| "loss": 0.1363, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 70, | |
| "total_flos": 9.906332448168346e+16, | |
| "train_loss": 0.3359420874289104, | |
| "train_runtime": 4370.2836, | |
| "train_samples_per_second": 1.602, | |
| "train_steps_per_second": 0.016 | |
| } | |
| ], | |
| "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": 9.906332448168346e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |