| { | |
| "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": 6.592547259478091, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 1.0519, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 6.567455308840202, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 1.0335, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 5.940088825094661, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 1.0137, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 2.8490940172517343, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.9749, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 4.051471478703239, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.9427, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 4.064845725783386, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.9398, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 4.123631763256305, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9153, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 3.0784568570852775, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.9034, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 2.2573484073375982, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.8591, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 1.5770602873647688, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.8049, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 1.276352557519139, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 0.7908, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 1.4262974839409701, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.7886, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 1.4161139727150824, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.7647, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 0.985317354399831, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.7351, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 1.1731937626326456, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.745, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 1.2342351925190727, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.7438, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 0.8586648305989024, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.7215, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 1.2623342706689424, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.7282, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 1.0216036051872144, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.7154, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.8625523662853701, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.6982, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 0.9240003298803581, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.7014, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 0.7872081733723811, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.6724, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 0.6179867079850143, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.6739, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 0.7504143540003096, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.6576, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.6238627830885334, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.6357, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 0.787718950541536, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.6485, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.5841746630896626, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.6373, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.6595519957236287, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.6327, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.5810919433489632, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.638, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.5598203359351309, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.6268, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 0.5081893908295881, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.6262, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 0.46842003845293695, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.5969, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 0.5411918223030173, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.5954, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.2640000000000002, | |
| "grad_norm": 0.4186008205264484, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.5805, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.5012713173626475, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.5929, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.456, | |
| "grad_norm": 0.42142495249376466, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.5865, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.552, | |
| "grad_norm": 0.4755105712649461, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.5861, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.648, | |
| "grad_norm": 0.46101341025428494, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.5732, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7439999999999998, | |
| "grad_norm": 0.43246687184135535, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.5584, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.39254399972143406, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.5548, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.936, | |
| "grad_norm": 0.3861973176797019, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.5655, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.032, | |
| "grad_norm": 0.42374098534776666, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.5633, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 0.4538854225702699, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.5378, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.224, | |
| "grad_norm": 0.3781458261925399, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.543, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.41511928586481533, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.5347, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.416, | |
| "grad_norm": 0.402421179277013, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.5447, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.5120000000000005, | |
| "grad_norm": 0.3556682564046602, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.5287, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 0.4086586929836952, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.5415, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.704, | |
| "grad_norm": 0.37123732327538245, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.5299, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.33105088139680716, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.54, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.896, | |
| "grad_norm": 0.34036827210692955, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.5145, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.992, | |
| "grad_norm": 0.37280888371104687, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.522, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.088, | |
| "grad_norm": 0.33200357438780415, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.5161, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.184, | |
| "grad_norm": 0.3468598712867419, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.5171, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.3155159225190127, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.497, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.376, | |
| "grad_norm": 0.2950796973774278, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.5036, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.4719999999999995, | |
| "grad_norm": 0.2771484456896418, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.5076, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.568, | |
| "grad_norm": 0.30202891361186884, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.5087, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.664, | |
| "grad_norm": 0.284433863235484, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.4892, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.28905567583309655, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.5286, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.856, | |
| "grad_norm": 0.30113786550921257, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.5088, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.952, | |
| "grad_norm": 0.29406826587431034, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.4946, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.048, | |
| "grad_norm": 0.2620114991759815, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.4894, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.144, | |
| "grad_norm": 0.2536132887815071, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.5078, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.2493726460319065, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.4851, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.336, | |
| "grad_norm": 0.24730699837443748, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.5004, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.432, | |
| "grad_norm": 0.24267186596083232, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.4937, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5280000000000005, | |
| "grad_norm": 0.23960620066529348, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.497, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.624, | |
| "grad_norm": 0.25488562503009676, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.4973, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.24026471856709594, | |
| "learning_rate": 0.0, | |
| "loss": 0.4974, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 70, | |
| "total_flos": 2.7617679424199066e+17, | |
| "train_loss": 0.639291496362005, | |
| "train_runtime": 11118.3141, | |
| "train_samples_per_second": 0.63, | |
| "train_steps_per_second": 0.006 | |
| } | |
| ], | |
| "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": 2.7617679424199066e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |