| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 325, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 0.2686200088875347, | |
| "learning_rate": 6.060606060606061e-06, | |
| "loss": 0.7782, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.21475453367529, | |
| "learning_rate": 3.0303030303030306e-05, | |
| "loss": 0.7797, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 0.2247471478195798, | |
| "learning_rate": 6.060606060606061e-05, | |
| "loss": 0.7851, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.20570321309918235, | |
| "learning_rate": 9.090909090909092e-05, | |
| "loss": 0.7854, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.18031349892436885, | |
| "learning_rate": 0.00012121212121212122, | |
| "loss": 0.7919, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.24146229252382978, | |
| "learning_rate": 0.00015151515151515152, | |
| "loss": 0.738, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.3320926970306046, | |
| "learning_rate": 0.00018181818181818183, | |
| "loss": 0.7418, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.19858557544056218, | |
| "learning_rate": 0.00019997685019798912, | |
| "loss": 0.764, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.19954091433241658, | |
| "learning_rate": 0.0001997165380022878, | |
| "loss": 0.7607, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.1889847467880647, | |
| "learning_rate": 0.000199167731989929, | |
| "loss": 0.7949, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.19722641297574658, | |
| "learning_rate": 0.0001983320199330545, | |
| "loss": 0.7642, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.19369172161622056, | |
| "learning_rate": 0.00019721181966290613, | |
| "loss": 0.7837, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.18633655144200606, | |
| "learning_rate": 0.00019581037207470382, | |
| "loss": 0.7569, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.17978045786081515, | |
| "learning_rate": 0.00019413173175128473, | |
| "loss": 0.7615, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.1946967703536159, | |
| "learning_rate": 0.00019218075523263104, | |
| "loss": 0.7537, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.1697959010297178, | |
| "learning_rate": 0.00018996308696522433, | |
| "loss": 0.7617, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.19424776053519632, | |
| "learning_rate": 0.00018748514297187648, | |
| "loss": 0.7995, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.17591012762345642, | |
| "learning_rate": 0.00018475409228928312, | |
| "loss": 0.754, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.18668198691171894, | |
| "learning_rate": 0.00018177783622700327, | |
| "loss": 0.7754, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.1741754965716253, | |
| "learning_rate": 0.00017856498550787144, | |
| "loss": 0.7923, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.18855993719083636, | |
| "learning_rate": 0.00017512483535597867, | |
| "loss": 0.7546, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.16677585411695353, | |
| "learning_rate": 0.00017146733860429612, | |
| "loss": 0.7752, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 0.2057091607658948, | |
| "learning_rate": 0.0001676030768997445, | |
| "loss": 0.745, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 0.18783062116028282, | |
| "learning_rate": 0.00016354323008901776, | |
| "loss": 0.7499, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 0.1625675859425024, | |
| "learning_rate": 0.00015929954387373103, | |
| "loss": 0.7347, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 0.19803022161416892, | |
| "learning_rate": 0.00015488429582847192, | |
| "loss": 0.791, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.19340769155849927, | |
| "learning_rate": 0.00015031025988006936, | |
| "loss": 0.7844, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 0.18701663817803796, | |
| "learning_rate": 0.00014559066935084588, | |
| "loss": 0.7883, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 0.16387851112042653, | |
| "learning_rate": 0.00014073917867277557, | |
| "loss": 0.7477, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 0.16958293060069174, | |
| "learning_rate": 0.0001357698238833126, | |
| "loss": 0.7577, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 0.17304291851159886, | |
| "learning_rate": 0.000130696982017182, | |
| "loss": 0.7349, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.20233311247327937, | |
| "learning_rate": 0.0001255353295116187, | |
| "loss": 0.7579, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 0.18036287897941758, | |
| "learning_rate": 0.00012029979974539234, | |
| "loss": 0.7428, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 0.18711131061340178, | |
| "learning_rate": 0.00011500553983446527, | |
| "loss": 0.7913, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.17301090514375908, | |
| "learning_rate": 0.00010966786680927874, | |
| "loss": 0.7278, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 0.19697238676167705, | |
| "learning_rate": 0.00010430222330045304, | |
| "loss": 0.7802, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.1688341794712563, | |
| "learning_rate": 9.892413286110886e-05, | |
| "loss": 0.7521, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 0.1770041663872631, | |
| "learning_rate": 9.354915505506839e-05, | |
| "loss": 0.7652, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 0.17975920055364886, | |
| "learning_rate": 8.81928404408726e-05, | |
| "loss": 0.7423, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.16596750546842118, | |
| "learning_rate": 8.287068558185225e-05, | |
| "loss": 0.785, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 0.17922768297335867, | |
| "learning_rate": 7.759808821241406e-05, | |
| "loss": 0.7469, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 0.16797159131863615, | |
| "learning_rate": 7.239030269025311e-05, | |
| "loss": 0.7434, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 0.17963390706599044, | |
| "learning_rate": 6.726239586337408e-05, | |
| "loss": 0.76, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 0.1632820830493435, | |
| "learning_rate": 6.22292034796035e-05, | |
| "loss": 0.753, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 0.1826122995064924, | |
| "learning_rate": 5.730528726470792e-05, | |
| "loss": 0.7681, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "grad_norm": 0.1613160734854796, | |
| "learning_rate": 5.2504892793295e-05, | |
| "loss": 0.743, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 0.18621725463530472, | |
| "learning_rate": 4.7841908274384616e-05, | |
| "loss": 0.7717, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.1744166288344059, | |
| "learning_rate": 4.332982437088825e-05, | |
| "loss": 0.7149, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 0.17659716102656, | |
| "learning_rate": 3.898169516924398e-05, | |
| "loss": 0.751, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.1715227056995353, | |
| "learning_rate": 3.4810100412128747e-05, | |
| "loss": 0.7561, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 0.17506626977459022, | |
| "learning_rate": 3.0827109103512643e-05, | |
| "loss": 0.7669, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 0.17815639423266896, | |
| "learning_rate": 2.7044244591351232e-05, | |
| "loss": 0.7589, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.1636846820858173, | |
| "learning_rate": 2.3472451228937253e-05, | |
| "loss": 0.7393, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 0.17100053615356214, | |
| "learning_rate": 2.0122062711363532e-05, | |
| "loss": 0.7764, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 0.17768328277125636, | |
| "learning_rate": 1.7002772178705716e-05, | |
| "loss": 0.7331, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 0.16988419995617995, | |
| "learning_rate": 1.4123604172419713e-05, | |
| "loss": 0.7719, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 0.18468128236342862, | |
| "learning_rate": 1.149288852608743e-05, | |
| "loss": 0.7558, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.16594356186499187, | |
| "learning_rate": 9.118236266049707e-06, | |
| "loss": 0.7309, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 0.1726455301629271, | |
| "learning_rate": 7.0065175916482095e-06, | |
| "loss": 0.76, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 0.1521864857478625, | |
| "learning_rate": 5.163841998782837e-06, | |
| "loss": 0.7642, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 0.1629366183790264, | |
| "learning_rate": 3.595540604290437e-06, | |
| "loss": 0.7578, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 0.17095198936108816, | |
| "learning_rate": 2.30615072228183e-06, | |
| "loss": 0.7332, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 0.17577071939294034, | |
| "learning_rate": 1.2994027370611173e-06, | |
| "loss": 0.7624, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 0.15277306413297514, | |
| "learning_rate": 5.782093106048159e-07, | |
| "loss": 0.7367, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "grad_norm": 0.17579866149327247, | |
| "learning_rate": 1.446569558255395e-07, | |
| "loss": 0.7431, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.16982851480438013, | |
| "learning_rate": 0.0, | |
| "loss": 0.749, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.7775599956512451, | |
| "eval_runtime": 47.5812, | |
| "eval_samples_per_second": 4.855, | |
| "eval_steps_per_second": 0.168, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 325, | |
| "total_flos": 4962652747988992.0, | |
| "train_loss": 0.7600007471671472, | |
| "train_runtime": 11976.0052, | |
| "train_samples_per_second": 1.736, | |
| "train_steps_per_second": 0.027 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 325, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 100, | |
| "total_flos": 4962652747988992.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |