| { | |
| "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": 6.456063172328566, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 1.0353, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 6.689620191479466, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 1.0543, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 5.937309490726284, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 1.0363, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 2.8406048640964, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.9809, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 4.0061919123179015, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.9086, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 4.203487007107819, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.9554, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 4.124004316023573, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9149, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 3.029781180420243, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.8903, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 2.2511238729024723, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.8456, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 1.6533144198168643, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.8179, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 1.262269228506302, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 0.7918, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 1.3193633842943835, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.7592, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 1.3888242168354306, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.7785, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 0.9657619090682279, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.7622, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 1.077266062640386, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.735, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 0.9176151147428974, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.7301, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 0.7742457520471991, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.7429, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 0.9478746648122993, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.7215, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 0.7399384465405956, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.7171, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.8154735736432236, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.6891, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 0.6221733254081043, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.6961, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 0.7933754546156063, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.649, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 0.6128147517657437, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.6771, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 0.7669016696341897, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.6587, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.5796966247202553, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.6357, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 0.7424326216776499, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.6431, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.46306860410480094, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.6341, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.6051227267124847, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.6476, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.5753231848210846, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.624, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.5272833129778086, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.6233, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 0.58216763773676, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.6252, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 0.5841409565638442, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.5995, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 0.5544017253080548, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.5774, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.2640000000000002, | |
| "grad_norm": 0.5030220536270247, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.5783, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.5494552136338695, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.5801, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.456, | |
| "grad_norm": 0.5134717170213757, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.585, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.552, | |
| "grad_norm": 0.4579135588857278, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.5745, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.648, | |
| "grad_norm": 0.45710699818927486, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.5644, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7439999999999998, | |
| "grad_norm": 0.3897602018157776, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.5688, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.45542942040682627, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.5636, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.936, | |
| "grad_norm": 0.4008931053291384, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.567, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.032, | |
| "grad_norm": 0.39983108289702124, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.5496, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 0.47926096410559416, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.533, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.224, | |
| "grad_norm": 0.3841507731419245, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.5171, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.41451120542191316, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.5225, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.416, | |
| "grad_norm": 0.47845747854627135, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.5357, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.5120000000000005, | |
| "grad_norm": 0.4513255851667966, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.5399, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 0.37833109565293876, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.5223, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.704, | |
| "grad_norm": 0.4262474433676995, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.532, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.356920052277616, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.5287, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.896, | |
| "grad_norm": 0.3336615935725844, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.5352, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.992, | |
| "grad_norm": 0.36728219621515973, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.5109, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.088, | |
| "grad_norm": 0.3626936220293368, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.4998, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.184, | |
| "grad_norm": 0.3237452361609368, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.5015, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.32897541207708086, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.4812, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.376, | |
| "grad_norm": 0.3302743116442111, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.5, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.4719999999999995, | |
| "grad_norm": 0.31341384372225634, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.5212, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.568, | |
| "grad_norm": 0.30010429291457386, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.4904, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.664, | |
| "grad_norm": 0.30726750197281294, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.4997, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.35312535310510246, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.5059, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.856, | |
| "grad_norm": 0.291873228894928, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.4886, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.952, | |
| "grad_norm": 0.2842981012605982, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.5207, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.048, | |
| "grad_norm": 0.30734278397409526, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.4788, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.144, | |
| "grad_norm": 0.26568650947302663, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.5147, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.25975925105687564, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.4976, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.336, | |
| "grad_norm": 0.25676316776284436, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.4994, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.432, | |
| "grad_norm": 0.2443198209753474, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.4974, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5280000000000005, | |
| "grad_norm": 0.27669024516016644, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.4767, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.624, | |
| "grad_norm": 0.2702475264212275, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.478, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.2872132385920433, | |
| "learning_rate": 0.0, | |
| "loss": 0.4762, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 70, | |
| "total_flos": 2.7768224651568742e+17, | |
| "train_loss": 0.6356262339012964, | |
| "train_runtime": 8792.7531, | |
| "train_samples_per_second": 0.796, | |
| "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": 2.7768224651568742e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |