| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.666666666666667, | |
| "eval_steps": 500, | |
| "global_step": 70, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09523809523809523, | |
| "grad_norm": 6.415903824560614, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 1.0361, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.19047619047619047, | |
| "grad_norm": 6.431761126160115, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 1.0203, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.2857142857142857, | |
| "grad_norm": 6.031595636100169, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 1.0252, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.38095238095238093, | |
| "grad_norm": 2.8425339393978306, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.9399, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.47619047619047616, | |
| "grad_norm": 4.108813074869778, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.9303, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.5714285714285714, | |
| "grad_norm": 4.283735043485786, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.9429, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.6666666666666666, | |
| "grad_norm": 4.202008376978637, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9108, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.7619047619047619, | |
| "grad_norm": 2.987285861727341, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.8734, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.8571428571428571, | |
| "grad_norm": 2.38499448366736, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.8219, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.9523809523809523, | |
| "grad_norm": 1.6498944196801038, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.8135, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.0476190476190477, | |
| "grad_norm": 2.126381162417613, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 1.1617, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.1428571428571428, | |
| "grad_norm": 1.4216840349744957, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.7806, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.2380952380952381, | |
| "grad_norm": 1.1126206823662455, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.7465, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3333333333333333, | |
| "grad_norm": 1.1060868329953828, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.741, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.4285714285714286, | |
| "grad_norm": 1.3480707841108062, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.7513, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.5238095238095237, | |
| "grad_norm": 1.128506631825107, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.6712, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.619047619047619, | |
| "grad_norm": 0.7943119639036299, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.7215, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "grad_norm": 1.0403036112949484, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.7001, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8095238095238095, | |
| "grad_norm": 0.8100999708913371, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.7002, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.9047619047619047, | |
| "grad_norm": 0.8357179181013383, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.7007, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.1261495486595832, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 1.0359, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.0952380952380953, | |
| "grad_norm": 0.7504139923001901, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.6654, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.1904761904761907, | |
| "grad_norm": 0.6447625619426562, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.641, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.2857142857142856, | |
| "grad_norm": 0.8300506967958257, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.6145, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.380952380952381, | |
| "grad_norm": 0.6225513224118171, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.6207, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.4761904761904763, | |
| "grad_norm": 0.7391175327947072, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.6362, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.571428571428571, | |
| "grad_norm": 0.5334576850613642, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.6348, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6666666666666665, | |
| "grad_norm": 0.5235703232925774, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.6093, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.761904761904762, | |
| "grad_norm": 0.642998375719684, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.6165, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 0.4658053830817156, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.6108, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.9523809523809526, | |
| "grad_norm": 0.5966923050382847, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.609, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.0476190476190474, | |
| "grad_norm": 0.7716838257337986, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.8791, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.142857142857143, | |
| "grad_norm": 0.587879200808507, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.5573, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.238095238095238, | |
| "grad_norm": 0.5583140647464201, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.6056, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.3333333333333335, | |
| "grad_norm": 0.5108631591598876, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.5801, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.4285714285714284, | |
| "grad_norm": 0.4689058367866831, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.5095, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.5238095238095237, | |
| "grad_norm": 0.5597549004472177, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.5514, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.619047619047619, | |
| "grad_norm": 0.5134898399930786, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.565, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7142857142857144, | |
| "grad_norm": 0.4223099277558807, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.5436, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.8095238095238093, | |
| "grad_norm": 0.4710520551064762, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.5537, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.9047619047619047, | |
| "grad_norm": 0.49065338745783726, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.5467, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.6548756574380363, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.8489, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.095238095238095, | |
| "grad_norm": 0.482649618346389, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.5117, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.190476190476191, | |
| "grad_norm": 0.45697394889696885, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.5139, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.285714285714286, | |
| "grad_norm": 0.4673884536822491, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.5227, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.380952380952381, | |
| "grad_norm": 0.36734233975527586, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.5081, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.476190476190476, | |
| "grad_norm": 0.41436622185111244, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.5099, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.571428571428571, | |
| "grad_norm": 0.43810236464798835, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.507, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.666666666666667, | |
| "grad_norm": 0.394087455831572, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.5036, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.761904761904762, | |
| "grad_norm": 0.4289117679669436, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.5092, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.857142857142857, | |
| "grad_norm": 0.4462542382162421, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.5083, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.9523809523809526, | |
| "grad_norm": 0.33369968059206156, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.5122, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.0476190476190474, | |
| "grad_norm": 0.7022648287433294, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.7209, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.142857142857143, | |
| "grad_norm": 0.38834718897982984, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.5082, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.238095238095238, | |
| "grad_norm": 0.3077809078195141, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.4518, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.333333333333333, | |
| "grad_norm": 0.33229212868146646, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.4791, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.428571428571429, | |
| "grad_norm": 0.35599517214464294, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.4981, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.523809523809524, | |
| "grad_norm": 0.33219002908305534, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.4833, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.619047619047619, | |
| "grad_norm": 0.32175034612592185, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.4504, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 0.2995559393640013, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.4814, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.809523809523809, | |
| "grad_norm": 0.31954673742524736, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.5159, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.904761904761905, | |
| "grad_norm": 0.28399758589547613, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.4504, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.505835968894034, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.7418, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.095238095238095, | |
| "grad_norm": 0.2919596742123474, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.4736, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.190476190476191, | |
| "grad_norm": 0.27090459847430504, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.4735, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.285714285714286, | |
| "grad_norm": 0.2486635722750014, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.4751, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.380952380952381, | |
| "grad_norm": 0.27374511923257594, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.4777, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.476190476190476, | |
| "grad_norm": 0.28562134036219766, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.4553, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.571428571428571, | |
| "grad_norm": 0.2589664217113908, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.4777, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.666666666666667, | |
| "grad_norm": 0.2555350772444153, | |
| "learning_rate": 0.0, | |
| "loss": 0.478, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.666666666666667, | |
| "step": 70, | |
| "total_flos": 1.9574749296748134e+17, | |
| "train_loss": 0.6460428876536233, | |
| "train_runtime": 2908.963, | |
| "train_samples_per_second": 2.406, | |
| "train_steps_per_second": 0.024 | |
| } | |
| ], | |
| "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": 1.9574749296748134e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |