| { | |
| "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": 7.183526298197059, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 1.2159, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 7.031646024711875, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 1.1803, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 6.726449054019057, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 1.2637, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 3.019251490604518, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 1.0997, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 4.372464279714231, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 1.134, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 4.668877413562089, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 1.1521, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 5.315985360539466, | |
| "learning_rate": 2e-05, | |
| "loss": 1.0869, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 3.531866863804273, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 1.0384, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 2.360792496559264, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.9982, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 2.0730886631612866, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.9549, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 2.0949143881823304, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 0.9849, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 1.4418707389397343, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.9482, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 1.4176378835481556, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.9402, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 1.213085102759749, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.9275, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 1.20171438446904, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.9148, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 1.1294101542052173, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.9246, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 1.0994307104874157, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.9171, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 1.020464421286651, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.9417, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 0.7681952849638046, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.8709, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.9490121535512666, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.9026, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 0.6572000695564753, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.8912, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 0.7890941812184483, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.8605, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 0.645276849098759, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.833, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 0.6944310244083031, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.8128, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.5769863100807223, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.8633, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 0.7467973758778769, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.8878, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.7020191157406843, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.7629, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.7102558998323939, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.8378, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 1.5903597029550902, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.795, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 1.2308988723660046, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.8061, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 1.8813596533220647, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.8583, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 0.8623655173556376, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.7933, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 276.41503463731976, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.8317, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.2640000000000002, | |
| "grad_norm": 0.6904839668045957, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.764, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.5664303765415496, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.7052, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.456, | |
| "grad_norm": 0.5220759676992732, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.7776, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.552, | |
| "grad_norm": 0.48786029357197297, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.7307, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.648, | |
| "grad_norm": 0.637056641485944, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.7632, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7439999999999998, | |
| "grad_norm": 0.5274190066086378, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.7752, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.4742218489863218, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.7477, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.936, | |
| "grad_norm": 0.4906516507530535, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.7471, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.032, | |
| "grad_norm": 0.42574380900812253, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.7555, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 0.4612817312885407, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.7426, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.224, | |
| "grad_norm": 0.4048890809208518, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.7229, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.5950686810700512, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.7084, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.416, | |
| "grad_norm": 0.4083958557164403, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.7176, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.5120000000000005, | |
| "grad_norm": 0.3889024722667518, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.6971, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 0.38627888537778465, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.6593, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.704, | |
| "grad_norm": 0.41107376175312915, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.7331, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.368851573498416, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.6982, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.896, | |
| "grad_norm": 0.3655873384272704, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.7143, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.992, | |
| "grad_norm": 0.36698174251614346, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.7085, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.088, | |
| "grad_norm": 0.3565085992843195, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.7289, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.184, | |
| "grad_norm": 0.33850857151098473, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.677, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.34277177833492384, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.6235, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.376, | |
| "grad_norm": 0.35409138917491984, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.6884, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.4719999999999995, | |
| "grad_norm": 0.3835765040879716, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.6943, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.568, | |
| "grad_norm": 0.3116079973081498, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.6462, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.664, | |
| "grad_norm": 0.31059743253395206, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.6899, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.3333135479564294, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.6851, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.856, | |
| "grad_norm": 0.3545171656487777, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.7027, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.952, | |
| "grad_norm": 0.34270948410594704, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.6706, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.048, | |
| "grad_norm": 0.313217720743749, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.693, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.144, | |
| "grad_norm": 0.31090294988670286, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.6914, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.3052177834188217, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.6366, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.336, | |
| "grad_norm": 0.2906275385092298, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.6899, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.432, | |
| "grad_norm": 0.2881426610713051, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.6732, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5280000000000005, | |
| "grad_norm": 0.2736825814044455, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.7035, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.624, | |
| "grad_norm": 0.2908985031551779, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.6699, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.2973945837536026, | |
| "learning_rate": 0.0, | |
| "loss": 0.7055, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 70, | |
| "total_flos": 2.0848443398710886e+17, | |
| "train_loss": 0.8195367446967534, | |
| "train_runtime": 7873.6613, | |
| "train_samples_per_second": 0.889, | |
| "train_steps_per_second": 0.009 | |
| } | |
| ], | |
| "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": 2.0848443398710886e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |