| { | |
| "best_metric": 0.010564341209828854, | |
| "best_model_checkpoint": "/home/paperspace/Data/models/relianceV2/llm3br256/checkpoint-145", | |
| "epoch": 4.957264957264957, | |
| "eval_steps": 5, | |
| "global_step": 145, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03418803418803419, | |
| "grad_norm": 0.14103831350803375, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.0642, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.06837606837606838, | |
| "grad_norm": 0.14193229377269745, | |
| "learning_rate": 1.3333333333333333e-05, | |
| "loss": 0.0655, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.10256410256410256, | |
| "grad_norm": 0.14674556255340576, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0667, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.13675213675213677, | |
| "grad_norm": 0.12595196068286896, | |
| "learning_rate": 2.6666666666666667e-05, | |
| "loss": 0.0607, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.17094017094017094, | |
| "grad_norm": 0.0885230153799057, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.0529, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.17094017094017094, | |
| "eval_loss": 0.04289048910140991, | |
| "eval_runtime": 19.6377, | |
| "eval_samples_per_second": 2.546, | |
| "eval_steps_per_second": 0.662, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.20512820512820512, | |
| "grad_norm": 0.06661484390497208, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0417, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.23931623931623933, | |
| "grad_norm": 0.08302908390760422, | |
| "learning_rate": 4.666666666666667e-05, | |
| "loss": 0.0405, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.27350427350427353, | |
| "grad_norm": 0.07343365997076035, | |
| "learning_rate": 5.333333333333333e-05, | |
| "loss": 0.0381, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.3076923076923077, | |
| "grad_norm": 0.05457036942243576, | |
| "learning_rate": 6e-05, | |
| "loss": 0.0378, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.3418803418803419, | |
| "grad_norm": 0.04456571117043495, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 0.0333, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.3418803418803419, | |
| "eval_loss": 0.031967516988515854, | |
| "eval_runtime": 20.2255, | |
| "eval_samples_per_second": 2.472, | |
| "eval_steps_per_second": 0.643, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.37606837606837606, | |
| "grad_norm": 0.04081442579627037, | |
| "learning_rate": 7.333333333333333e-05, | |
| "loss": 0.03, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.41025641025641024, | |
| "grad_norm": 0.03535589948296547, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0308, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.4444444444444444, | |
| "grad_norm": 0.03202543035149574, | |
| "learning_rate": 8.666666666666667e-05, | |
| "loss": 0.0273, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.47863247863247865, | |
| "grad_norm": 0.031043237075209618, | |
| "learning_rate": 9.333333333333334e-05, | |
| "loss": 0.0232, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.5128205128205128, | |
| "grad_norm": 0.027209093794226646, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0247, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.5128205128205128, | |
| "eval_loss": 0.025287916883826256, | |
| "eval_runtime": 20.2717, | |
| "eval_samples_per_second": 2.466, | |
| "eval_steps_per_second": 0.641, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.5470085470085471, | |
| "grad_norm": 0.0389866903424263, | |
| "learning_rate": 9.998540070400966e-05, | |
| "loss": 0.0256, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.5811965811965812, | |
| "grad_norm": 0.02687215991318226, | |
| "learning_rate": 9.994161134161634e-05, | |
| "loss": 0.0215, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.6153846153846154, | |
| "grad_norm": 0.026792502030730247, | |
| "learning_rate": 9.986865748457457e-05, | |
| "loss": 0.0268, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.6495726495726496, | |
| "grad_norm": 0.028072144836187363, | |
| "learning_rate": 9.976658173588244e-05, | |
| "loss": 0.0216, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.6837606837606838, | |
| "grad_norm": 0.024400506168603897, | |
| "learning_rate": 9.96354437049027e-05, | |
| "loss": 0.0228, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.6837606837606838, | |
| "eval_loss": 0.02183925360441208, | |
| "eval_runtime": 20.441, | |
| "eval_samples_per_second": 2.446, | |
| "eval_steps_per_second": 0.636, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.717948717948718, | |
| "grad_norm": 0.023090992122888565, | |
| "learning_rate": 9.947531997255256e-05, | |
| "loss": 0.0211, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.7521367521367521, | |
| "grad_norm": 0.021273907274007797, | |
| "learning_rate": 9.928630404658255e-05, | |
| "loss": 0.0191, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.7863247863247863, | |
| "grad_norm": 0.021417921409010887, | |
| "learning_rate": 9.906850630697068e-05, | |
| "loss": 0.02, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.8205128205128205, | |
| "grad_norm": 0.02200082130730152, | |
| "learning_rate": 9.882205394146361e-05, | |
| "loss": 0.0232, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.8547008547008547, | |
| "grad_norm": 0.02200504206120968, | |
| "learning_rate": 9.85470908713026e-05, | |
| "loss": 0.0196, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.8547008547008547, | |
| "eval_loss": 0.019536755979061127, | |
| "eval_runtime": 20.402, | |
| "eval_samples_per_second": 2.451, | |
| "eval_steps_per_second": 0.637, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.8888888888888888, | |
| "grad_norm": 0.02056378312408924, | |
| "learning_rate": 9.824377766717759e-05, | |
| "loss": 0.0208, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.9230769230769231, | |
| "grad_norm": 0.018094003200531006, | |
| "learning_rate": 9.791229145545831e-05, | |
| "loss": 0.0167, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.9572649572649573, | |
| "grad_norm": 0.017437463626265526, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.0178, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.9914529914529915, | |
| "grad_norm": 0.016863780096173286, | |
| "learning_rate": 9.716559066288715e-05, | |
| "loss": 0.0163, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.0256410256410255, | |
| "grad_norm": 0.043561145663261414, | |
| "learning_rate": 9.675081213427076e-05, | |
| "loss": 0.0342, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.0256410256410255, | |
| "eval_loss": 0.018172133713960648, | |
| "eval_runtime": 20.3342, | |
| "eval_samples_per_second": 2.459, | |
| "eval_steps_per_second": 0.639, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.0598290598290598, | |
| "grad_norm": 0.024272920563817024, | |
| "learning_rate": 9.630873244788883e-05, | |
| "loss": 0.016, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.0940170940170941, | |
| "grad_norm": 0.025347765535116196, | |
| "learning_rate": 9.583960976582913e-05, | |
| "loss": 0.016, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.1282051282051282, | |
| "grad_norm": 0.026027293875813484, | |
| "learning_rate": 9.534371804252728e-05, | |
| "loss": 0.0178, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.1623931623931625, | |
| "grad_norm": 0.021461905911564827, | |
| "learning_rate": 9.482134686478519e-05, | |
| "loss": 0.0166, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.1965811965811965, | |
| "grad_norm": 0.025416361168026924, | |
| "learning_rate": 9.42728012826605e-05, | |
| "loss": 0.014, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.1965811965811965, | |
| "eval_loss": 0.016842123121023178, | |
| "eval_runtime": 20.4307, | |
| "eval_samples_per_second": 2.447, | |
| "eval_steps_per_second": 0.636, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.2307692307692308, | |
| "grad_norm": 0.021278902888298035, | |
| "learning_rate": 9.36984016313259e-05, | |
| "loss": 0.0147, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.264957264957265, | |
| "grad_norm": 0.023418180644512177, | |
| "learning_rate": 9.309848334400246e-05, | |
| "loss": 0.0155, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.2991452991452992, | |
| "grad_norm": 0.021139057353138924, | |
| "learning_rate": 9.247339675607605e-05, | |
| "loss": 0.015, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.3333333333333333, | |
| "grad_norm": 0.021048279479146004, | |
| "learning_rate": 9.182350690051133e-05, | |
| "loss": 0.0148, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.3675213675213675, | |
| "grad_norm": 0.022858263924717903, | |
| "learning_rate": 9.114919329468282e-05, | |
| "loss": 0.0142, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.3675213675213675, | |
| "eval_loss": 0.01600724086165428, | |
| "eval_runtime": 20.1138, | |
| "eval_samples_per_second": 2.486, | |
| "eval_steps_per_second": 0.646, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.4017094017094016, | |
| "grad_norm": 0.020670941099524498, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.0157, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.435897435897436, | |
| "grad_norm": 0.01937521994113922, | |
| "learning_rate": 8.972888398568772e-05, | |
| "loss": 0.0181, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.4700854700854702, | |
| "grad_norm": 0.02165701799094677, | |
| "learning_rate": 8.898371770316111e-05, | |
| "loss": 0.014, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.5042735042735043, | |
| "grad_norm": 0.01963304542005062, | |
| "learning_rate": 8.821578602729242e-05, | |
| "loss": 0.0154, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 0.016389695927500725, | |
| "learning_rate": 8.742553740855506e-05, | |
| "loss": 0.0133, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "eval_loss": 0.014650734141469002, | |
| "eval_runtime": 20.2255, | |
| "eval_samples_per_second": 2.472, | |
| "eval_steps_per_second": 0.643, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.5726495726495726, | |
| "grad_norm": 0.02393350377678871, | |
| "learning_rate": 8.661343332988869e-05, | |
| "loss": 0.015, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.606837606837607, | |
| "grad_norm": 0.01642770506441593, | |
| "learning_rate": 8.577994803720606e-05, | |
| "loss": 0.0149, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.641025641025641, | |
| "grad_norm": 0.018430177122354507, | |
| "learning_rate": 8.492556826244687e-05, | |
| "loss": 0.0142, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.6752136752136753, | |
| "grad_norm": 0.021527277305722237, | |
| "learning_rate": 8.405079293933986e-05, | |
| "loss": 0.0125, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.7094017094017095, | |
| "grad_norm": 0.01738244667649269, | |
| "learning_rate": 8.315613291203976e-05, | |
| "loss": 0.0122, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.7094017094017095, | |
| "eval_loss": 0.013878260739147663, | |
| "eval_runtime": 20.266, | |
| "eval_samples_per_second": 2.467, | |
| "eval_steps_per_second": 0.641, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.7435897435897436, | |
| "grad_norm": 0.016872745007276535, | |
| "learning_rate": 8.224211063680853e-05, | |
| "loss": 0.0136, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 1.7777777777777777, | |
| "grad_norm": 0.01582753099501133, | |
| "learning_rate": 8.130925987691569e-05, | |
| "loss": 0.0125, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.811965811965812, | |
| "grad_norm": 0.01693882793188095, | |
| "learning_rate": 8.035812539093557e-05, | |
| "loss": 0.0105, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 1.8461538461538463, | |
| "grad_norm": 0.01629294827580452, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.0122, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.8803418803418803, | |
| "grad_norm": 0.02483530156314373, | |
| "learning_rate": 7.840323733655778e-05, | |
| "loss": 0.0136, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.8803418803418803, | |
| "eval_loss": 0.013426054269075394, | |
| "eval_runtime": 20.4254, | |
| "eval_samples_per_second": 2.448, | |
| "eval_steps_per_second": 0.636, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.9145299145299144, | |
| "grad_norm": 0.017920587211847305, | |
| "learning_rate": 7.740062536773352e-05, | |
| "loss": 0.0121, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.9487179487179487, | |
| "grad_norm": 0.018872473388910294, | |
| "learning_rate": 7.638201220530665e-05, | |
| "loss": 0.0124, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.982905982905983, | |
| "grad_norm": 0.018264025449752808, | |
| "learning_rate": 7.534799269067953e-05, | |
| "loss": 0.0125, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 2.017094017094017, | |
| "grad_norm": 0.042249832302331924, | |
| "learning_rate": 7.42991706621303e-05, | |
| "loss": 0.0197, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 2.051282051282051, | |
| "grad_norm": 0.025319905951619148, | |
| "learning_rate": 7.323615860218843e-05, | |
| "loss": 0.011, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.051282051282051, | |
| "eval_loss": 0.01329517550766468, | |
| "eval_runtime": 20.8923, | |
| "eval_samples_per_second": 2.393, | |
| "eval_steps_per_second": 0.622, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.0854700854700856, | |
| "grad_norm": 0.01969628967344761, | |
| "learning_rate": 7.215957727996207e-05, | |
| "loss": 0.0107, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 2.1196581196581197, | |
| "grad_norm": 0.021322010084986687, | |
| "learning_rate": 7.107005538862646e-05, | |
| "loss": 0.0107, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 2.1538461538461537, | |
| "grad_norm": 0.015889568254351616, | |
| "learning_rate": 6.996822917828477e-05, | |
| "loss": 0.0115, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 2.1880341880341883, | |
| "grad_norm": 0.014772447757422924, | |
| "learning_rate": 6.885474208441603e-05, | |
| "loss": 0.0105, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 2.2222222222222223, | |
| "grad_norm": 0.02105395495891571, | |
| "learning_rate": 6.773024435212678e-05, | |
| "loss": 0.0098, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.2222222222222223, | |
| "eval_loss": 0.012847676873207092, | |
| "eval_runtime": 20.8651, | |
| "eval_samples_per_second": 2.396, | |
| "eval_steps_per_second": 0.623, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.2564102564102564, | |
| "grad_norm": 0.017252441495656967, | |
| "learning_rate": 6.659539265642643e-05, | |
| "loss": 0.0105, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 2.2905982905982905, | |
| "grad_norm": 0.015464423224329948, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.0083, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 2.324786324786325, | |
| "grad_norm": 0.022285865619778633, | |
| "learning_rate": 6.429728391993446e-05, | |
| "loss": 0.0116, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.358974358974359, | |
| "grad_norm": 0.022020753473043442, | |
| "learning_rate": 6.313536890992935e-05, | |
| "loss": 0.0091, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 2.393162393162393, | |
| "grad_norm": 0.022549916058778763, | |
| "learning_rate": 6.19657832143779e-05, | |
| "loss": 0.011, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.393162393162393, | |
| "eval_loss": 0.012613310478627682, | |
| "eval_runtime": 20.423, | |
| "eval_samples_per_second": 2.448, | |
| "eval_steps_per_second": 0.637, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.427350427350427, | |
| "grad_norm": 0.021075539290905, | |
| "learning_rate": 6.078920983839031e-05, | |
| "loss": 0.0105, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 2.4615384615384617, | |
| "grad_norm": 0.02389822155237198, | |
| "learning_rate": 5.960633586768543e-05, | |
| "loss": 0.0105, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 2.4957264957264957, | |
| "grad_norm": 0.01924612931907177, | |
| "learning_rate": 5.841785206735192e-05, | |
| "loss": 0.0106, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 2.52991452991453, | |
| "grad_norm": 0.01965448632836342, | |
| "learning_rate": 5.7224452478461064e-05, | |
| "loss": 0.0108, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 2.564102564102564, | |
| "grad_norm": 0.015871981158852577, | |
| "learning_rate": 5.602683401276615e-05, | |
| "loss": 0.0087, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 2.564102564102564, | |
| "eval_loss": 0.012018020264804363, | |
| "eval_runtime": 20.4309, | |
| "eval_samples_per_second": 2.447, | |
| "eval_steps_per_second": 0.636, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 2.5982905982905984, | |
| "grad_norm": 0.017729448154568672, | |
| "learning_rate": 5.482569604572576e-05, | |
| "loss": 0.0085, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 2.6324786324786325, | |
| "grad_norm": 0.019757023081183434, | |
| "learning_rate": 5.3621740008088126e-05, | |
| "loss": 0.0111, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 2.6666666666666665, | |
| "grad_norm": 0.016380244866013527, | |
| "learning_rate": 5.2415668976275355e-05, | |
| "loss": 0.0081, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 2.700854700854701, | |
| "grad_norm": 0.017829637974500656, | |
| "learning_rate": 5.1208187261806615e-05, | |
| "loss": 0.0112, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 2.735042735042735, | |
| "grad_norm": 0.019623294472694397, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0086, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.735042735042735, | |
| "eval_loss": 0.011721721850335598, | |
| "eval_runtime": 20.2095, | |
| "eval_samples_per_second": 2.474, | |
| "eval_steps_per_second": 0.643, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.769230769230769, | |
| "grad_norm": 0.01674460805952549, | |
| "learning_rate": 4.87918127381934e-05, | |
| "loss": 0.009, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 2.8034188034188032, | |
| "grad_norm": 0.01912359707057476, | |
| "learning_rate": 4.758433102372466e-05, | |
| "loss": 0.01, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 2.8376068376068377, | |
| "grad_norm": 0.021485494449734688, | |
| "learning_rate": 4.6378259991911886e-05, | |
| "loss": 0.008, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 2.871794871794872, | |
| "grad_norm": 0.02604834921658039, | |
| "learning_rate": 4.5174303954274244e-05, | |
| "loss": 0.0093, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 2.905982905982906, | |
| "grad_norm": 0.01696549728512764, | |
| "learning_rate": 4.397316598723385e-05, | |
| "loss": 0.0091, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 2.905982905982906, | |
| "eval_loss": 0.01138822827488184, | |
| "eval_runtime": 20.1788, | |
| "eval_samples_per_second": 2.478, | |
| "eval_steps_per_second": 0.644, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 2.9401709401709404, | |
| "grad_norm": 0.02111894078552723, | |
| "learning_rate": 4.277554752153895e-05, | |
| "loss": 0.0098, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 2.9743589743589745, | |
| "grad_norm": 0.021606622263789177, | |
| "learning_rate": 4.1582147932648074e-05, | |
| "loss": 0.01, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 3.0085470085470085, | |
| "grad_norm": 0.04893641173839569, | |
| "learning_rate": 4.039366413231458e-05, | |
| "loss": 0.0134, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 3.0427350427350426, | |
| "grad_norm": 0.0206940695643425, | |
| "learning_rate": 3.92107901616097e-05, | |
| "loss": 0.0079, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 0.030171874910593033, | |
| "learning_rate": 3.803421678562213e-05, | |
| "loss": 0.009, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "eval_loss": 0.011335253715515137, | |
| "eval_runtime": 20.4524, | |
| "eval_samples_per_second": 2.445, | |
| "eval_steps_per_second": 0.636, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.111111111111111, | |
| "grad_norm": 0.02197062224149704, | |
| "learning_rate": 3.6864631090070655e-05, | |
| "loss": 0.0077, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 3.1452991452991452, | |
| "grad_norm": 0.018409984186291695, | |
| "learning_rate": 3.570271608006555e-05, | |
| "loss": 0.0076, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 3.1794871794871793, | |
| "grad_norm": 0.02173592895269394, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.007, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 3.213675213675214, | |
| "grad_norm": 0.030453743413090706, | |
| "learning_rate": 3.340460734357359e-05, | |
| "loss": 0.0072, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 3.247863247863248, | |
| "grad_norm": 0.01577071100473404, | |
| "learning_rate": 3.226975564787322e-05, | |
| "loss": 0.0062, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 3.247863247863248, | |
| "eval_loss": 0.011275053024291992, | |
| "eval_runtime": 11.7759, | |
| "eval_samples_per_second": 4.246, | |
| "eval_steps_per_second": 1.104, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 3.282051282051282, | |
| "grad_norm": 0.020528519526124, | |
| "learning_rate": 3.114525791558398e-05, | |
| "loss": 0.0077, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 3.316239316239316, | |
| "grad_norm": 0.028798261657357216, | |
| "learning_rate": 3.003177082171523e-05, | |
| "loss": 0.0076, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 3.3504273504273505, | |
| "grad_norm": 0.023175662383437157, | |
| "learning_rate": 2.8929944611373554e-05, | |
| "loss": 0.0072, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 3.3846153846153846, | |
| "grad_norm": 0.020334038883447647, | |
| "learning_rate": 2.784042272003794e-05, | |
| "loss": 0.0069, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 3.4188034188034186, | |
| "grad_norm": 0.024305541068315506, | |
| "learning_rate": 2.6763841397811573e-05, | |
| "loss": 0.0069, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.4188034188034186, | |
| "eval_loss": 0.011122054420411587, | |
| "eval_runtime": 9.713, | |
| "eval_samples_per_second": 5.148, | |
| "eval_steps_per_second": 1.338, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.452991452991453, | |
| "grad_norm": 0.022912999615073204, | |
| "learning_rate": 2.57008293378697e-05, | |
| "loss": 0.0094, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 3.4871794871794872, | |
| "grad_norm": 0.021239478141069412, | |
| "learning_rate": 2.4652007309320498e-05, | |
| "loss": 0.0064, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 3.5213675213675213, | |
| "grad_norm": 0.017512433230876923, | |
| "learning_rate": 2.361798779469336e-05, | |
| "loss": 0.0074, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 3.5555555555555554, | |
| "grad_norm": 0.018754741176962852, | |
| "learning_rate": 2.259937463226651e-05, | |
| "loss": 0.0063, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 3.58974358974359, | |
| "grad_norm": 0.019357949495315552, | |
| "learning_rate": 2.1596762663442218e-05, | |
| "loss": 0.0071, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 3.58974358974359, | |
| "eval_loss": 0.01076994277536869, | |
| "eval_runtime": 9.7069, | |
| "eval_samples_per_second": 5.151, | |
| "eval_steps_per_second": 1.339, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 3.623931623931624, | |
| "grad_norm": 0.02121426723897457, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.0078, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 3.658119658119658, | |
| "grad_norm": 0.020560307428240776, | |
| "learning_rate": 1.9641874609064443e-05, | |
| "loss": 0.0079, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 3.6923076923076925, | |
| "grad_norm": 0.024254970252513885, | |
| "learning_rate": 1.8690740123084316e-05, | |
| "loss": 0.0067, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 3.7264957264957266, | |
| "grad_norm": 0.020139144733548164, | |
| "learning_rate": 1.7757889363191483e-05, | |
| "loss": 0.0066, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 3.7606837606837606, | |
| "grad_norm": 0.01883375644683838, | |
| "learning_rate": 1.684386708796025e-05, | |
| "loss": 0.006, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 3.7606837606837606, | |
| "eval_loss": 0.010681088082492352, | |
| "eval_runtime": 9.7636, | |
| "eval_samples_per_second": 5.121, | |
| "eval_steps_per_second": 1.331, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 3.7948717948717947, | |
| "grad_norm": 0.017693698406219482, | |
| "learning_rate": 1.5949207060660138e-05, | |
| "loss": 0.0061, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 3.8290598290598292, | |
| "grad_norm": 0.020463695749640465, | |
| "learning_rate": 1.5074431737553157e-05, | |
| "loss": 0.0072, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 3.8632478632478633, | |
| "grad_norm": 0.020722465589642525, | |
| "learning_rate": 1.422005196279395e-05, | |
| "loss": 0.0091, | |
| "step": 113 | |
| }, | |
| { | |
| "epoch": 3.8974358974358974, | |
| "grad_norm": 0.020328683778643608, | |
| "learning_rate": 1.338656667011134e-05, | |
| "loss": 0.0071, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 3.931623931623932, | |
| "grad_norm": 0.0204046331346035, | |
| "learning_rate": 1.257446259144494e-05, | |
| "loss": 0.0065, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 3.931623931623932, | |
| "eval_loss": 0.010614048689603806, | |
| "eval_runtime": 9.7105, | |
| "eval_samples_per_second": 5.149, | |
| "eval_steps_per_second": 1.339, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 3.965811965811966, | |
| "grad_norm": 0.018496442586183548, | |
| "learning_rate": 1.178421397270758e-05, | |
| "loss": 0.0059, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.04243512079119682, | |
| "learning_rate": 1.1016282296838887e-05, | |
| "loss": 0.0121, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 4.034188034188034, | |
| "grad_norm": 0.01636839285492897, | |
| "learning_rate": 1.0271116014312293e-05, | |
| "loss": 0.005, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 4.068376068376068, | |
| "grad_norm": 0.016252394765615463, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.0053, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 4.102564102564102, | |
| "grad_norm": 0.016149073839187622, | |
| "learning_rate": 8.850806705317183e-06, | |
| "loss": 0.0076, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.102564102564102, | |
| "eval_loss": 0.010566945187747478, | |
| "eval_runtime": 9.7295, | |
| "eval_samples_per_second": 5.139, | |
| "eval_steps_per_second": 1.336, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.136752136752137, | |
| "grad_norm": 0.01633269712328911, | |
| "learning_rate": 8.176493099488663e-06, | |
| "loss": 0.0072, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 4.170940170940171, | |
| "grad_norm": 0.018178582191467285, | |
| "learning_rate": 7.526603243923957e-06, | |
| "loss": 0.0061, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 4.205128205128205, | |
| "grad_norm": 0.017663611099123955, | |
| "learning_rate": 6.901516655997536e-06, | |
| "loss": 0.006, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 4.239316239316239, | |
| "grad_norm": 0.01672632433474064, | |
| "learning_rate": 6.301598368674105e-06, | |
| "loss": 0.0059, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 4.273504273504273, | |
| "grad_norm": 0.01625417172908783, | |
| "learning_rate": 5.727198717339511e-06, | |
| "loss": 0.0057, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 4.273504273504273, | |
| "eval_loss": 0.010598141700029373, | |
| "eval_runtime": 9.7399, | |
| "eval_samples_per_second": 5.133, | |
| "eval_steps_per_second": 1.335, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 4.3076923076923075, | |
| "grad_norm": 0.016820693388581276, | |
| "learning_rate": 5.178653135214812e-06, | |
| "loss": 0.0051, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 4.3418803418803416, | |
| "grad_norm": 0.017715787515044212, | |
| "learning_rate": 4.65628195747273e-06, | |
| "loss": 0.0058, | |
| "step": 127 | |
| }, | |
| { | |
| "epoch": 4.3760683760683765, | |
| "grad_norm": 0.01836553029716015, | |
| "learning_rate": 4.16039023417088e-06, | |
| "loss": 0.0051, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 4.410256410256411, | |
| "grad_norm": 0.02160499058663845, | |
| "learning_rate": 3.691267552111183e-06, | |
| "loss": 0.0054, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 4.444444444444445, | |
| "grad_norm": 0.019846314564347267, | |
| "learning_rate": 3.249187865729264e-06, | |
| "loss": 0.0071, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 4.444444444444445, | |
| "eval_loss": 0.010587843134999275, | |
| "eval_runtime": 9.707, | |
| "eval_samples_per_second": 5.151, | |
| "eval_steps_per_second": 1.339, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 4.478632478632479, | |
| "grad_norm": 0.01771419309079647, | |
| "learning_rate": 2.8344093371128424e-06, | |
| "loss": 0.005, | |
| "step": 131 | |
| }, | |
| { | |
| "epoch": 4.512820512820513, | |
| "grad_norm": 0.015574207529425621, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 0.0048, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 4.547008547008547, | |
| "grad_norm": 0.01756260357797146, | |
| "learning_rate": 2.087708544541689e-06, | |
| "loss": 0.0047, | |
| "step": 133 | |
| }, | |
| { | |
| "epoch": 4.581196581196581, | |
| "grad_norm": 0.02206473983824253, | |
| "learning_rate": 1.7562223328224325e-06, | |
| "loss": 0.0053, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 4.615384615384615, | |
| "grad_norm": 0.017151422798633575, | |
| "learning_rate": 1.4529091286973995e-06, | |
| "loss": 0.0055, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 4.615384615384615, | |
| "eval_loss": 0.010594615712761879, | |
| "eval_runtime": 9.7136, | |
| "eval_samples_per_second": 5.147, | |
| "eval_steps_per_second": 1.338, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 4.64957264957265, | |
| "grad_norm": 0.018389733508229256, | |
| "learning_rate": 1.1779460585363944e-06, | |
| "loss": 0.0053, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 4.683760683760684, | |
| "grad_norm": 0.016087768599390984, | |
| "learning_rate": 9.314936930293283e-07, | |
| "loss": 0.0048, | |
| "step": 137 | |
| }, | |
| { | |
| "epoch": 4.717948717948718, | |
| "grad_norm": 0.019610892981290817, | |
| "learning_rate": 7.136959534174592e-07, | |
| "loss": 0.0056, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 4.752136752136752, | |
| "grad_norm": 0.018921220675110817, | |
| "learning_rate": 5.246800274474439e-07, | |
| "loss": 0.0065, | |
| "step": 139 | |
| }, | |
| { | |
| "epoch": 4.786324786324786, | |
| "grad_norm": 0.020553983747959137, | |
| "learning_rate": 3.6455629509730136e-07, | |
| "loss": 0.0053, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 4.786324786324786, | |
| "eval_loss": 0.010573416948318481, | |
| "eval_runtime": 9.7174, | |
| "eval_samples_per_second": 5.145, | |
| "eval_steps_per_second": 1.338, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 4.82051282051282, | |
| "grad_norm": 0.015205741859972477, | |
| "learning_rate": 2.334182641175686e-07, | |
| "loss": 0.0038, | |
| "step": 141 | |
| }, | |
| { | |
| "epoch": 4.854700854700854, | |
| "grad_norm": 0.017062170431017876, | |
| "learning_rate": 1.3134251542544774e-07, | |
| "loss": 0.006, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 4.888888888888889, | |
| "grad_norm": 0.01831440068781376, | |
| "learning_rate": 5.838865838366792e-08, | |
| "loss": 0.0063, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 4.923076923076923, | |
| "grad_norm": 0.019888985902071, | |
| "learning_rate": 1.4599295990352924e-08, | |
| "loss": 0.0068, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 4.957264957264957, | |
| "grad_norm": 0.01834353245794773, | |
| "learning_rate": 0.0, | |
| "loss": 0.0074, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 4.957264957264957, | |
| "eval_loss": 0.010564341209828854, | |
| "eval_runtime": 9.7196, | |
| "eval_samples_per_second": 5.144, | |
| "eval_steps_per_second": 1.338, | |
| "step": 145 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 145, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 50, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.332736073022669e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |