| { | |
| "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": 6.504410456047468, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 1.0916, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 6.615938735764493, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 1.0821, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 6.214217921378788, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 1.0489, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 2.635084377615703, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.989, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 4.719230750157174, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.9696, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 4.907344720711667, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.9739, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 4.920011518238829, | |
| "learning_rate": 2e-05, | |
| "loss": 0.8694, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 4.224074601936279, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.9052, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 3.9803873111692094, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.8579, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 2.941525554582962, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.8244, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 2.5598294321149098, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 0.8047, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 2.2847739297049072, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.779, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 1.7790316901994792, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.7461, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 1.8157022786267045, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.7333, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 1.2259210953288842, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.7313, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 1.1403026880897884, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.7092, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 1.131993491557917, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.7022, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 0.7869712180154732, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.6912, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 0.9875156286513475, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.6894, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.9052966286287072, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.6854, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 1.0264832061021716, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.6995, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 1.0776736206194137, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.5837, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 0.8625955047069487, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.5745, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 0.8518714197682477, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.5618, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.9755111982110702, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.5291, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 0.7811345213430506, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.5331, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.793162568696382, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.5432, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.7304847514692294, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.5508, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.7583405854452997, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.5452, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.7084558435571406, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.5103, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 0.7330019478891183, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.5167, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 0.7158422565661753, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.4624, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 0.644546071565366, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.4226, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.2640000000000002, | |
| "grad_norm": 0.659473025422639, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.4306, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.7965943853116153, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.4443, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.456, | |
| "grad_norm": 0.6480730171495981, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.3998, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.552, | |
| "grad_norm": 0.7428237561066205, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.3922, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.648, | |
| "grad_norm": 0.6079575918493917, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.4302, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7439999999999998, | |
| "grad_norm": 0.7000704381409165, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.4083, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.6681438169490255, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.4012, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.936, | |
| "grad_norm": 0.6229919175764043, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.3906, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.032, | |
| "grad_norm": 0.5831622864960437, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.3681, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 0.6375352620367634, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.3641, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.224, | |
| "grad_norm": 0.5734053009914873, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.3145, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.5014720277885562, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.3419, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.416, | |
| "grad_norm": 0.6548961319605631, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.3779, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.5120000000000005, | |
| "grad_norm": 0.6490365489244189, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.3171, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 0.524484395525688, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.3163, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.704, | |
| "grad_norm": 0.5267265650841149, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.3272, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.512627176895878, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.2966, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.896, | |
| "grad_norm": 0.4924517861158664, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.3029, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.992, | |
| "grad_norm": 0.4259287978233709, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.3279, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.088, | |
| "grad_norm": 0.5200073283715068, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.2863, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.184, | |
| "grad_norm": 0.4692629551954398, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.2844, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.4198687322595458, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.2997, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.376, | |
| "grad_norm": 0.4070492563315593, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.2537, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.4719999999999995, | |
| "grad_norm": 0.47471218536527954, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.2911, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.568, | |
| "grad_norm": 0.47739422922038743, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.276, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.664, | |
| "grad_norm": 0.4251325666312027, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.2847, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.4014461359314857, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.2653, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.856, | |
| "grad_norm": 0.38226345917240023, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.2916, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.952, | |
| "grad_norm": 0.38500061345656394, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.3008, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.048, | |
| "grad_norm": 0.394650855919485, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.2991, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.144, | |
| "grad_norm": 0.4298668762623011, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.2926, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.3546155283248144, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.2762, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.336, | |
| "grad_norm": 0.3835475116128774, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.2842, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.432, | |
| "grad_norm": 0.35812643025190777, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.2726, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5280000000000005, | |
| "grad_norm": 0.4083299304064863, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.2599, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.624, | |
| "grad_norm": 0.370273974800934, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.2711, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.4115438169992172, | |
| "learning_rate": 0.0, | |
| "loss": 0.2514, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 70, | |
| "total_flos": 1.215241503291474e+17, | |
| "train_loss": 0.5129862508603504, | |
| "train_runtime": 8669.0336, | |
| "train_samples_per_second": 0.807, | |
| "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": 1.215241503291474e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |