| { | |
| "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": 5.944960466387328, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 1.0114, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 6.086483553324464, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 1.041, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 5.99718694277905, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 1.0925, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 2.5945150141338624, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.9667, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 3.691322070793699, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.9125, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 3.636564757900175, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.9275, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 3.777217276184943, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9128, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 2.801895386381312, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.8821, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 2.10629151324274, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.8571, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 1.6440589900064753, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.8137, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 2.092646032889902, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 1.3113, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 1.1499358617208464, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.6879, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 1.2388038395776648, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.7442, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 1.0595966899070013, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.7407, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 1.0162689746872489, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.6901, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 1.2220238283984195, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.7608, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 0.8945443299192337, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.6895, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 1.0666561139043083, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.7521, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 0.9597100466790252, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.6698, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.7769867887536602, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.6473, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 1.288167245486445, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 1.1523, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 0.8422032392080524, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.6157, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 0.7703640645225037, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.6178, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 0.963217183539902, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.6317, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 2.42970396082646, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.6027, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 0.8164650857124783, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.5533, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.7786368384278178, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.6066, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.710769695741534, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.5678, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.8865322858182426, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.6325, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.5849363222112286, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.5373, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 0.9767844704922458, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.6994, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 0.86164608116822, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.7936, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 0.8237193547271372, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.5078, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.2640000000000002, | |
| "grad_norm": 0.8100977134743738, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.5623, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.5667085900879959, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.4179, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.456, | |
| "grad_norm": 1.1019994059262113, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.5445, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.552, | |
| "grad_norm": 0.7091810014433372, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.4995, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.648, | |
| "grad_norm": 0.9405628300626622, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.5807, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7439999999999998, | |
| "grad_norm": 0.6516410423639595, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.4443, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.8414105089077197, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.4994, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.936, | |
| "grad_norm": 0.666993922442764, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.5195, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.032, | |
| "grad_norm": 0.9682206402164324, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.7055, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 0.5090195286609236, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.4386, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.224, | |
| "grad_norm": 0.6237639543839937, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.437, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.546011697031731, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.4205, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.416, | |
| "grad_norm": 0.5830602559492782, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.4537, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.5120000000000005, | |
| "grad_norm": 0.5867223908669553, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.4246, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 0.6031989296649094, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.4397, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.704, | |
| "grad_norm": 0.5125385478443895, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.4246, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.5367122362937156, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.4292, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.896, | |
| "grad_norm": 0.4887001636196693, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.4398, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.992, | |
| "grad_norm": 0.7739002780453024, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.621, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.088, | |
| "grad_norm": 0.5224586264775329, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.4077, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.184, | |
| "grad_norm": 0.47639037276254115, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.3825, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.4390100164868234, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.4008, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.376, | |
| "grad_norm": 0.42262001081899037, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.3816, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.4719999999999995, | |
| "grad_norm": 0.43954542690884474, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.405, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.568, | |
| "grad_norm": 0.38418311432799357, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.3762, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.664, | |
| "grad_norm": 0.4074795238463937, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.3877, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.3790306148786251, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.3881, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.856, | |
| "grad_norm": 0.4030792590172934, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.3721, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.952, | |
| "grad_norm": 0.3809741315217957, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.3866, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.048, | |
| "grad_norm": 0.7607656624613088, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.6044, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.144, | |
| "grad_norm": 0.34745927753316835, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.336, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.3543815281347, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.35, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.336, | |
| "grad_norm": 0.3632710267355952, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.3763, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.432, | |
| "grad_norm": 0.36488026498237486, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.3805, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5280000000000005, | |
| "grad_norm": 0.358041962170842, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.3826, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.624, | |
| "grad_norm": 0.3414392418565718, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.3613, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.34777568542354104, | |
| "learning_rate": 0.0, | |
| "loss": 0.3879, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 70, | |
| "total_flos": 1.2820678594422374e+17, | |
| "train_loss": 0.5999843043940408, | |
| "train_runtime": 5632.331, | |
| "train_samples_per_second": 1.243, | |
| "train_steps_per_second": 0.012 | |
| } | |
| ], | |
| "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": 1.2820678594422374e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |