| { | |
| "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": 6.451047433495054, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 1.0466, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 6.890186408504931, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 1.0807, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 6.252060135307135, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 1.0755, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 2.6119371716205775, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.9502, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 4.6624418761736255, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.9743, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 4.825828175600524, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.9333, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 5.145684175929915, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9137, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 4.176913602611626, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.8766, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 3.651935910791593, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.8852, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 3.0396179735618887, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.827, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 2.7273164791101165, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 0.7995, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 2.2703614643602097, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.7696, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 1.9564585855172414, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.7688, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 1.6311574409934144, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.7243, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 1.1788645000726892, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.7241, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 1.47834766670541, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.7258, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 1.159011627283112, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.6674, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 1.0286519572111426, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.6931, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 0.9917441325180145, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.7073, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.8450348990379875, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.6679, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 1.019389347282369, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.6792, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 1.014524243445132, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.5971, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 0.9241922223915863, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.569, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 1.116124475774628, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.5527, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 1.0891305033509877, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.5579, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 0.9270747158541157, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.5475, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.8276099962652581, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.5401, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.7728127512847315, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.5181, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.9423511875314863, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.5141, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.76767033376674, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.508, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 0.7416323463245816, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.5201, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 0.8883416388126079, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.4799, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 0.7427450224085967, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.4337, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.2640000000000002, | |
| "grad_norm": 0.6605968653736728, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.4528, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.8149408216548297, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.4128, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.456, | |
| "grad_norm": 0.5558446483647723, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.3951, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.552, | |
| "grad_norm": 0.6776106797896108, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.4351, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.648, | |
| "grad_norm": 0.6918634957854057, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.3822, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7439999999999998, | |
| "grad_norm": 0.5623747097109255, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.4186, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.6320337238119385, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.396, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.936, | |
| "grad_norm": 0.6529909418203007, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.378, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.032, | |
| "grad_norm": 0.5635621756163047, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.3786, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 0.6660643638374503, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.3213, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.224, | |
| "grad_norm": 0.615145922586785, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.3146, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.5036450380615838, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.3027, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.416, | |
| "grad_norm": 0.6902192876367795, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.3089, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.5120000000000005, | |
| "grad_norm": 0.6981782845597782, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.3455, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 0.45688399908844185, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.3249, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.704, | |
| "grad_norm": 0.4817781158802941, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.3273, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.5420445586206913, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.3502, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.896, | |
| "grad_norm": 0.49936534561475665, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.3097, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.992, | |
| "grad_norm": 0.48643502560797053, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.3394, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.088, | |
| "grad_norm": 0.5051269543519935, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.2872, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.184, | |
| "grad_norm": 0.44242884595366866, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.2509, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.40949587969552276, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.2942, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.376, | |
| "grad_norm": 0.5184194423944949, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.2709, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.4719999999999995, | |
| "grad_norm": 0.4869492017110009, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.2611, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.568, | |
| "grad_norm": 0.4874499873662536, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.3068, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.664, | |
| "grad_norm": 0.40739662462312576, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.2957, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.4175190778810535, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.3184, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.856, | |
| "grad_norm": 0.38181524082595636, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.2488, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.952, | |
| "grad_norm": 0.3739777871765848, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.2965, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.048, | |
| "grad_norm": 0.41487790241808353, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.2678, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.144, | |
| "grad_norm": 0.4788502822962079, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.3017, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.39660756885219817, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.2882, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.336, | |
| "grad_norm": 0.3775925054864233, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.2521, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.432, | |
| "grad_norm": 0.3781496026359709, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.227, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5280000000000005, | |
| "grad_norm": 0.37746654494685233, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.2574, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.624, | |
| "grad_norm": 0.34584772547163234, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.3008, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.38710660627219096, | |
| "learning_rate": 0.0, | |
| "loss": 0.2726, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 70, | |
| "total_flos": 1.1873544016022733e+17, | |
| "train_loss": 0.5102894621235984, | |
| "train_runtime": 8358.5661, | |
| "train_samples_per_second": 0.837, | |
| "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": 1.1873544016022733e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |