| { | |
| "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": 6.280240805311102, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 0.866, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 6.679765431372183, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 0.8978, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 5.568228626143876, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 0.8151, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 2.626836059600364, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.7911, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 4.2325083339894425, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.7936, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 4.14089518478729, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.722, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 4.6440153905502175, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6895, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 3.2504936505069266, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.7104, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 2.7557631918299363, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.6641, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 2.210178505578961, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.6747, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 2.894504417886847, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 1.0109, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 1.1465010565493337, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.5785, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 1.199215219278408, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.6205, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 1.139232574152057, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.5816, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 0.8966331206331358, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.6112, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 0.9293694463729928, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.5568, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 0.9024711149274084, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.5767, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 0.9352246506778062, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.5486, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 0.8542809019966106, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.6001, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.6815754098805157, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.556, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 1.05881385118882, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.8103, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 0.7520031069950387, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.5266, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 0.6020947621638394, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.4972, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 0.7956938474149279, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.5222, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.6299170128054989, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.5045, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 0.6440503263568208, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.449, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.6351525902445594, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.5056, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.6361386932230259, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.498, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.6015303341911363, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.4543, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.5741925675314585, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.4929, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 0.5912855589895262, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.5402, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 0.7534856367552475, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.6607, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 0.4756477447162196, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.435, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.2640000000000002, | |
| "grad_norm": 0.5077043914811501, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.417, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.4966683306012897, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.4425, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.456, | |
| "grad_norm": 0.489546713248074, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.435, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.552, | |
| "grad_norm": 0.5135858554333199, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.439, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.648, | |
| "grad_norm": 0.47561577961714774, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.4146, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7439999999999998, | |
| "grad_norm": 0.47687835578387133, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.4204, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.431833297681947, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.4202, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.936, | |
| "grad_norm": 0.36051517375303277, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.3878, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.032, | |
| "grad_norm": 0.7759779497895979, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.6768, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 0.39343051442168503, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.3567, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.224, | |
| "grad_norm": 0.5587259976426712, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.3923, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.4135675021073055, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.3507, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.416, | |
| "grad_norm": 0.417934102862686, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.4036, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.5120000000000005, | |
| "grad_norm": 0.5015810367294763, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.3818, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 0.38098499597801916, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.3629, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.704, | |
| "grad_norm": 0.37029103578862615, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.3656, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.4132018045253673, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.3815, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.896, | |
| "grad_norm": 0.37594316887176193, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.3639, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.992, | |
| "grad_norm": 0.5253929889381646, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.5241, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.088, | |
| "grad_norm": 0.45699216826422856, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.4079, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.184, | |
| "grad_norm": 0.3416049853297944, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.3375, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.3380484573759369, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.3396, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.376, | |
| "grad_norm": 0.34027208441762447, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.3446, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.4719999999999995, | |
| "grad_norm": 0.3479117834499629, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.3252, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.568, | |
| "grad_norm": 0.3369205213237811, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.3357, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.664, | |
| "grad_norm": 0.3303146470866375, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.3356, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.32565062777582454, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.3603, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.856, | |
| "grad_norm": 0.30259668377127963, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.3471, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.952, | |
| "grad_norm": 0.3149573720853517, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.3593, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.048, | |
| "grad_norm": 0.5095882312843404, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.5618, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.144, | |
| "grad_norm": 0.27314963023544453, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.3101, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.271948324952245, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.3383, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.336, | |
| "grad_norm": 0.28573733481949803, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.3654, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.432, | |
| "grad_norm": 0.26580546390531196, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.3379, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5280000000000005, | |
| "grad_norm": 0.29529380179175113, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.3448, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.624, | |
| "grad_norm": 0.26453724776444787, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.2931, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.29323489579445655, | |
| "learning_rate": 0.0, | |
| "loss": 0.3613, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 70, | |
| "total_flos": 1.6215994151836058e+17, | |
| "train_loss": 0.5014790096453258, | |
| "train_runtime": 6471.8029, | |
| "train_samples_per_second": 1.082, | |
| "train_steps_per_second": 0.011 | |
| } | |
| ], | |
| "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": 1.6215994151836058e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |