| { | |
| "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": 6.696238353268114, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 1.0741, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 6.956192634327631, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 1.1354, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 6.079672922357755, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 1.0572, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 2.655842863349976, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.9717, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 4.6962008139151115, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 1.046, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 5.136558323915918, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.9998, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 5.189521642013073, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9791, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 3.938754870626044, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.9266, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 3.173124320200447, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.8921, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 3.0834787678972657, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.8964, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 3.5067078025428478, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 1.3485, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 1.7421410225843332, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.7922, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 1.7308247248134734, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.7967, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 1.3864464475766891, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.7539, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 1.2115033126994978, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.6981, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 1.359735011039239, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.7069, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 1.000048796454394, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.6688, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 0.9968125559644502, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.7604, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 0.9326664531110149, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.7584, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.8155592033544385, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.6776, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 1.3917521423230623, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 1.1475, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 1.0796691824651705, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.6127, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 5.532254123320559, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.5727, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 0.984495159742896, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.5873, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.8384740237837615, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.5552, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 1.0824796539805643, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.5277, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.7909507911704086, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.5507, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.710512095243773, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.5058, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.7688650221264791, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.4986, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.823376245193563, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.5527, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 1.0555491386969642, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.6294, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 1.1297644777498677, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.6523, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 0.7242454321251177, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.41, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.2640000000000002, | |
| "grad_norm": 0.7574983916517163, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.4649, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.8016900223552272, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.3895, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.456, | |
| "grad_norm": 0.7661683895769823, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.3931, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.552, | |
| "grad_norm": 0.6880928891314685, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.3983, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.648, | |
| "grad_norm": 0.6653517552243607, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.3718, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7439999999999998, | |
| "grad_norm": 0.6474575023652429, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.3645, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.7324987274228084, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.3952, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.936, | |
| "grad_norm": 0.7077903055497389, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.3919, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.032, | |
| "grad_norm": 1.2042804487430379, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.6276, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 0.5916329507961869, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.3009, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.224, | |
| "grad_norm": 0.6803369545615578, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.2784, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.6064755901207506, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.3214, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.416, | |
| "grad_norm": 0.6854924361841132, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.2703, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.5120000000000005, | |
| "grad_norm": 0.6461631226303043, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.2515, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 0.5576409784057318, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.279, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.704, | |
| "grad_norm": 0.5765521947770055, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.2749, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.6234565422014346, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.2756, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.896, | |
| "grad_norm": 0.5394043151498752, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.2566, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.992, | |
| "grad_norm": 0.873687326408933, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.4153, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.088, | |
| "grad_norm": 0.6980925561344368, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.2395, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.184, | |
| "grad_norm": 0.5896958547950758, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.2364, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.4890358978928442, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.235, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.376, | |
| "grad_norm": 0.517877196277932, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.2039, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.4719999999999995, | |
| "grad_norm": 0.5802443516889878, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.2112, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.568, | |
| "grad_norm": 0.6149216955187231, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.2151, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.664, | |
| "grad_norm": 0.561603182888772, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.2074, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.577509910756153, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.2497, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.856, | |
| "grad_norm": 0.463776673859254, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.1987, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.952, | |
| "grad_norm": 0.4538477930931362, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.2069, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.048, | |
| "grad_norm": 0.7883757393756762, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.3269, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.144, | |
| "grad_norm": 0.5147680033558916, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.2441, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.4609762943678798, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.1867, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.336, | |
| "grad_norm": 0.4770564059576249, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.2144, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.432, | |
| "grad_norm": 0.435751695262066, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.1744, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5280000000000005, | |
| "grad_norm": 0.477966755080849, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.1935, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.624, | |
| "grad_norm": 0.46295789437247015, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.1982, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.48109862071917014, | |
| "learning_rate": 0.0, | |
| "loss": 0.1973, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 70, | |
| "total_flos": 6.475553493431091e+16, | |
| "train_loss": 0.5200352900794574, | |
| "train_runtime": 2858.772, | |
| "train_samples_per_second": 2.449, | |
| "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": 6.475553493431091e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |