| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.75, | |
| "eval_steps": 500, | |
| "global_step": 70, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09375, | |
| "grad_norm": 2.875503705356323, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 0.7458, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.1875, | |
| "grad_norm": 2.990735042222842, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 0.7602, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.28125, | |
| "grad_norm": 2.8743554915409346, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 0.7519, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.375, | |
| "grad_norm": 1.4772155527540136, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.7153, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.46875, | |
| "grad_norm": 1.3260714657658668, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.6684, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.5625, | |
| "grad_norm": 1.1210151382840188, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.64, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.65625, | |
| "grad_norm": 1.928672362983636, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6327, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 1.3492877961036636, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.6231, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.84375, | |
| "grad_norm": 1.682740189167653, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.549, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.9375, | |
| "grad_norm": 1.3222526769254588, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.5613, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.0625, | |
| "grad_norm": 1.7776044627317216, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 0.9283, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.15625, | |
| "grad_norm": 0.6787199178640781, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.5032, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 0.7267722871293905, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.4411, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.34375, | |
| "grad_norm": 0.8081353104439796, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.5585, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.4375, | |
| "grad_norm": 0.6320827316278724, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.5122, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.53125, | |
| "grad_norm": 0.6454145407007941, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.472, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.625, | |
| "grad_norm": 0.7650252540496532, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.4393, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.71875, | |
| "grad_norm": 0.5636187172161755, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.5286, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8125, | |
| "grad_norm": 0.4876522873094857, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.446, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.90625, | |
| "grad_norm": 0.5614039576077123, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.4645, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.03125, | |
| "grad_norm": 1.011631880110753, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.7493, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.125, | |
| "grad_norm": 0.48981037970039826, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.4312, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.21875, | |
| "grad_norm": 0.5824671327559001, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.3837, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.3125, | |
| "grad_norm": 0.4951836648461806, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.3636, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.40625, | |
| "grad_norm": 0.40746509032687167, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.3929, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 0.47270781275959445, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.4309, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.59375, | |
| "grad_norm": 0.4139970793315771, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.347, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6875, | |
| "grad_norm": 0.4094525531542567, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.3898, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.78125, | |
| "grad_norm": 0.3609227071911589, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.3793, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.875, | |
| "grad_norm": 0.3475046450366041, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.3784, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.96875, | |
| "grad_norm": 0.6690800228387159, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.652, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.09375, | |
| "grad_norm": 0.44950641151876647, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.3481, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.1875, | |
| "grad_norm": 0.3848478424704702, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.3106, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.28125, | |
| "grad_norm": 0.5347567465033009, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.3268, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.375, | |
| "grad_norm": 0.4072495441359064, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.3324, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.46875, | |
| "grad_norm": 0.38962221521765955, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.3175, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.5625, | |
| "grad_norm": 0.3788865699519135, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.3257, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.65625, | |
| "grad_norm": 0.395721175839029, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.306, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "grad_norm": 0.33584385177429055, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.2622, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84375, | |
| "grad_norm": 0.39573379970895023, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.3478, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.9375, | |
| "grad_norm": 0.35397292743924824, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.3309, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.0625, | |
| "grad_norm": 0.6876995002147458, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.4723, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.15625, | |
| "grad_norm": 0.4197974469543892, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.2464, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "grad_norm": 0.38253306452841973, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.228, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.34375, | |
| "grad_norm": 0.3901250987784833, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.2946, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.4375, | |
| "grad_norm": 0.36777969276301314, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.2977, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.53125, | |
| "grad_norm": 0.3425452852047255, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.2405, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.625, | |
| "grad_norm": 0.34222956283167716, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.2637, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.71875, | |
| "grad_norm": 0.3010985646978744, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.2559, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8125, | |
| "grad_norm": 0.3204500290045904, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.2402, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.90625, | |
| "grad_norm": 0.43376115329479936, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.2634, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 5.03125, | |
| "grad_norm": 0.5277252915121167, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.3428, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.125, | |
| "grad_norm": 0.3242694914181531, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.2751, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.21875, | |
| "grad_norm": 0.2629237100248874, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.206, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.3125, | |
| "grad_norm": 0.26869870078975083, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.208, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.40625, | |
| "grad_norm": 0.2946292151795919, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.176, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.5, | |
| "grad_norm": 0.35978658604512964, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.247, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.59375, | |
| "grad_norm": 0.2694338117234328, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.2397, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.6875, | |
| "grad_norm": 0.23603362724985336, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.2557, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.78125, | |
| "grad_norm": 0.22274653612755244, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.2265, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.875, | |
| "grad_norm": 0.24217335349194297, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.2466, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.96875, | |
| "grad_norm": 0.48048696089483417, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.3496, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.09375, | |
| "grad_norm": 0.2936135705752043, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.1815, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.1875, | |
| "grad_norm": 0.2767171386917678, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.2077, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.28125, | |
| "grad_norm": 0.2345675305906572, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.2032, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.375, | |
| "grad_norm": 0.23553355433831719, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.2458, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.46875, | |
| "grad_norm": 0.20099324393208673, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.2298, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5625, | |
| "grad_norm": 0.21234850892293564, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.2243, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.65625, | |
| "grad_norm": 0.2120148750108375, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.2301, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.75, | |
| "grad_norm": 0.2753841782457882, | |
| "learning_rate": 0.0, | |
| "loss": 0.2001, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.75, | |
| "step": 70, | |
| "total_flos": 1.0066919896508006e+17, | |
| "train_loss": 0.3906510761805943, | |
| "train_runtime": 9044.8257, | |
| "train_samples_per_second": 0.774, | |
| "train_steps_per_second": 0.008 | |
| } | |
| ], | |
| "logging_steps": 1.0, | |
| "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": 1.0066919896508006e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |