jonasknobloch's picture
Initial commit
b4320f3
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.0,
"eval_steps": 10000,
"global_step": 267952,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"learning_rate": 4.9906699707410284e-05,
"loss": 7.0772,
"step": 500
},
{
"epoch": 0.0,
"learning_rate": 4.9813399414820566e-05,
"loss": 6.1824,
"step": 1000
},
{
"epoch": 0.01,
"learning_rate": 4.972009912223085e-05,
"loss": 5.8409,
"step": 1500
},
{
"epoch": 0.01,
"learning_rate": 4.962679882964113e-05,
"loss": 5.5985,
"step": 2000
},
{
"epoch": 0.01,
"learning_rate": 4.953349853705141e-05,
"loss": 5.403,
"step": 2500
},
{
"epoch": 0.01,
"learning_rate": 4.94401982444617e-05,
"loss": 5.2549,
"step": 3000
},
{
"epoch": 0.01,
"learning_rate": 4.9346897951871976e-05,
"loss": 5.1225,
"step": 3500
},
{
"epoch": 0.01,
"learning_rate": 4.9253597659282264e-05,
"loss": 4.9967,
"step": 4000
},
{
"epoch": 0.02,
"learning_rate": 4.916029736669254e-05,
"loss": 4.8795,
"step": 4500
},
{
"epoch": 0.02,
"learning_rate": 4.906699707410283e-05,
"loss": 4.7848,
"step": 5000
},
{
"epoch": 0.02,
"learning_rate": 4.897369678151311e-05,
"loss": 4.6995,
"step": 5500
},
{
"epoch": 0.02,
"learning_rate": 4.888039648892339e-05,
"loss": 4.6346,
"step": 6000
},
{
"epoch": 0.02,
"learning_rate": 4.8787096196333674e-05,
"loss": 4.5872,
"step": 6500
},
{
"epoch": 0.03,
"learning_rate": 4.8693795903743955e-05,
"loss": 4.521,
"step": 7000
},
{
"epoch": 0.03,
"learning_rate": 4.860049561115424e-05,
"loss": 4.4799,
"step": 7500
},
{
"epoch": 0.03,
"learning_rate": 4.850719531856452e-05,
"loss": 4.4502,
"step": 8000
},
{
"epoch": 0.03,
"learning_rate": 4.84138950259748e-05,
"loss": 4.4183,
"step": 8500
},
{
"epoch": 0.03,
"learning_rate": 4.832059473338509e-05,
"loss": 4.3762,
"step": 9000
},
{
"epoch": 0.04,
"learning_rate": 4.8227294440795365e-05,
"loss": 4.3529,
"step": 9500
},
{
"epoch": 0.04,
"learning_rate": 4.8133994148205653e-05,
"loss": 4.3227,
"step": 10000
},
{
"epoch": 0.04,
"eval_accuracy": 0.3160632599821231,
"eval_loss": 4.226778984069824,
"eval_runtime": 6188.29,
"eval_samples_per_second": 72.909,
"eval_steps_per_second": 1.139,
"step": 10000
},
{
"epoch": 0.04,
"learning_rate": 4.804069385561593e-05,
"loss": 4.2976,
"step": 10500
},
{
"epoch": 0.04,
"learning_rate": 4.794739356302622e-05,
"loss": 4.2731,
"step": 11000
},
{
"epoch": 0.04,
"learning_rate": 4.78540932704365e-05,
"loss": 4.248,
"step": 11500
},
{
"epoch": 0.04,
"learning_rate": 4.776079297784678e-05,
"loss": 4.2336,
"step": 12000
},
{
"epoch": 0.05,
"learning_rate": 4.766749268525706e-05,
"loss": 4.2144,
"step": 12500
},
{
"epoch": 0.05,
"learning_rate": 4.7574192392667345e-05,
"loss": 4.1986,
"step": 13000
},
{
"epoch": 0.05,
"learning_rate": 4.7480892100077627e-05,
"loss": 4.1854,
"step": 13500
},
{
"epoch": 0.05,
"learning_rate": 4.7387591807487915e-05,
"loss": 4.1669,
"step": 14000
},
{
"epoch": 0.05,
"learning_rate": 4.729429151489819e-05,
"loss": 4.1441,
"step": 14500
},
{
"epoch": 0.06,
"learning_rate": 4.720099122230848e-05,
"loss": 4.1442,
"step": 15000
},
{
"epoch": 0.06,
"learning_rate": 4.7107690929718754e-05,
"loss": 4.1226,
"step": 15500
},
{
"epoch": 0.06,
"learning_rate": 4.701439063712904e-05,
"loss": 4.1112,
"step": 16000
},
{
"epoch": 0.06,
"learning_rate": 4.692109034453932e-05,
"loss": 4.1063,
"step": 16500
},
{
"epoch": 0.06,
"learning_rate": 4.6827790051949607e-05,
"loss": 4.0869,
"step": 17000
},
{
"epoch": 0.07,
"learning_rate": 4.673448975935989e-05,
"loss": 4.0732,
"step": 17500
},
{
"epoch": 0.07,
"learning_rate": 4.664118946677017e-05,
"loss": 4.0716,
"step": 18000
},
{
"epoch": 0.07,
"learning_rate": 4.654788917418045e-05,
"loss": 4.053,
"step": 18500
},
{
"epoch": 0.07,
"learning_rate": 4.6454588881590734e-05,
"loss": 4.0442,
"step": 19000
},
{
"epoch": 0.07,
"learning_rate": 4.6361288589001016e-05,
"loss": 4.035,
"step": 19500
},
{
"epoch": 0.07,
"learning_rate": 4.6267988296411305e-05,
"loss": 4.0305,
"step": 20000
},
{
"epoch": 0.07,
"eval_accuracy": 0.33925145807254103,
"eval_loss": 3.945496082305908,
"eval_runtime": 6149.9279,
"eval_samples_per_second": 73.364,
"eval_steps_per_second": 1.146,
"step": 20000
},
{
"epoch": 0.08,
"learning_rate": 4.617468800382158e-05,
"loss": 4.0218,
"step": 20500
},
{
"epoch": 0.08,
"learning_rate": 4.608138771123187e-05,
"loss": 4.0064,
"step": 21000
},
{
"epoch": 0.08,
"learning_rate": 4.5988087418642143e-05,
"loss": 4.0014,
"step": 21500
},
{
"epoch": 0.08,
"learning_rate": 4.589478712605243e-05,
"loss": 3.993,
"step": 22000
},
{
"epoch": 0.08,
"learning_rate": 4.5801486833462714e-05,
"loss": 3.9845,
"step": 22500
},
{
"epoch": 0.09,
"learning_rate": 4.5708186540872996e-05,
"loss": 3.9814,
"step": 23000
},
{
"epoch": 0.09,
"learning_rate": 4.561488624828328e-05,
"loss": 3.9689,
"step": 23500
},
{
"epoch": 0.09,
"learning_rate": 4.552158595569356e-05,
"loss": 3.9648,
"step": 24000
},
{
"epoch": 0.09,
"learning_rate": 4.542828566310384e-05,
"loss": 3.9599,
"step": 24500
},
{
"epoch": 0.09,
"learning_rate": 4.533498537051412e-05,
"loss": 3.9514,
"step": 25000
},
{
"epoch": 0.1,
"learning_rate": 4.5241685077924405e-05,
"loss": 3.9456,
"step": 25500
},
{
"epoch": 0.1,
"learning_rate": 4.5148384785334694e-05,
"loss": 3.9438,
"step": 26000
},
{
"epoch": 0.1,
"learning_rate": 4.505508449274497e-05,
"loss": 3.9315,
"step": 26500
},
{
"epoch": 0.1,
"learning_rate": 4.496178420015526e-05,
"loss": 3.9294,
"step": 27000
},
{
"epoch": 0.1,
"learning_rate": 4.486848390756553e-05,
"loss": 3.9137,
"step": 27500
},
{
"epoch": 0.1,
"learning_rate": 4.477518361497582e-05,
"loss": 3.9075,
"step": 28000
},
{
"epoch": 0.11,
"learning_rate": 4.46818833223861e-05,
"loss": 3.9069,
"step": 28500
},
{
"epoch": 0.11,
"learning_rate": 4.4588583029796385e-05,
"loss": 3.9039,
"step": 29000
},
{
"epoch": 0.11,
"learning_rate": 4.449528273720667e-05,
"loss": 3.9033,
"step": 29500
},
{
"epoch": 0.11,
"learning_rate": 4.440198244461695e-05,
"loss": 3.8916,
"step": 30000
},
{
"epoch": 0.11,
"eval_accuracy": 0.3501727973835296,
"eval_loss": 3.819415807723999,
"eval_runtime": 6177.9797,
"eval_samples_per_second": 73.031,
"eval_steps_per_second": 1.141,
"step": 30000
},
{
"epoch": 0.11,
"learning_rate": 4.430868215202723e-05,
"loss": 3.8876,
"step": 30500
},
{
"epoch": 0.12,
"learning_rate": 4.421538185943751e-05,
"loss": 3.8857,
"step": 31000
},
{
"epoch": 0.12,
"learning_rate": 4.4122081566847794e-05,
"loss": 3.8756,
"step": 31500
},
{
"epoch": 0.12,
"learning_rate": 4.4028781274258076e-05,
"loss": 3.871,
"step": 32000
},
{
"epoch": 0.12,
"learning_rate": 4.393548098166836e-05,
"loss": 3.8784,
"step": 32500
},
{
"epoch": 0.12,
"learning_rate": 4.384218068907865e-05,
"loss": 3.8657,
"step": 33000
},
{
"epoch": 0.13,
"learning_rate": 4.374888039648893e-05,
"loss": 3.8643,
"step": 33500
},
{
"epoch": 0.13,
"learning_rate": 4.365558010389921e-05,
"loss": 3.8518,
"step": 34000
},
{
"epoch": 0.13,
"learning_rate": 4.356227981130949e-05,
"loss": 3.8578,
"step": 34500
},
{
"epoch": 0.13,
"learning_rate": 4.3468979518719774e-05,
"loss": 3.8479,
"step": 35000
},
{
"epoch": 0.13,
"learning_rate": 4.3375679226130056e-05,
"loss": 3.846,
"step": 35500
},
{
"epoch": 0.13,
"learning_rate": 4.328237893354034e-05,
"loss": 3.8396,
"step": 36000
},
{
"epoch": 0.14,
"learning_rate": 4.318907864095062e-05,
"loss": 3.8401,
"step": 36500
},
{
"epoch": 0.14,
"learning_rate": 4.30957783483609e-05,
"loss": 3.8269,
"step": 37000
},
{
"epoch": 0.14,
"learning_rate": 4.3002478055771184e-05,
"loss": 3.8236,
"step": 37500
},
{
"epoch": 0.14,
"learning_rate": 4.2909177763181466e-05,
"loss": 3.8246,
"step": 38000
},
{
"epoch": 0.14,
"learning_rate": 4.281587747059175e-05,
"loss": 3.8236,
"step": 38500
},
{
"epoch": 0.15,
"learning_rate": 4.272257717800203e-05,
"loss": 3.8148,
"step": 39000
},
{
"epoch": 0.15,
"learning_rate": 4.262927688541232e-05,
"loss": 3.8033,
"step": 39500
},
{
"epoch": 0.15,
"learning_rate": 4.253597659282259e-05,
"loss": 3.8104,
"step": 40000
},
{
"epoch": 0.15,
"eval_accuracy": 0.3580380029125215,
"eval_loss": 3.7339699268341064,
"eval_runtime": 6179.879,
"eval_samples_per_second": 73.009,
"eval_steps_per_second": 1.141,
"step": 40000
},
{
"epoch": 0.15,
"learning_rate": 4.244267630023288e-05,
"loss": 3.809,
"step": 40500
},
{
"epoch": 0.15,
"learning_rate": 4.234937600764316e-05,
"loss": 3.8172,
"step": 41000
},
{
"epoch": 0.15,
"learning_rate": 4.2256075715053446e-05,
"loss": 3.8018,
"step": 41500
},
{
"epoch": 0.16,
"learning_rate": 4.216277542246373e-05,
"loss": 3.7956,
"step": 42000
},
{
"epoch": 0.16,
"learning_rate": 4.206947512987401e-05,
"loss": 3.7959,
"step": 42500
},
{
"epoch": 0.16,
"learning_rate": 4.197617483728429e-05,
"loss": 3.7884,
"step": 43000
},
{
"epoch": 0.16,
"learning_rate": 4.188287454469457e-05,
"loss": 3.788,
"step": 43500
},
{
"epoch": 0.16,
"learning_rate": 4.1789574252104855e-05,
"loss": 3.7795,
"step": 44000
},
{
"epoch": 0.17,
"learning_rate": 4.169627395951514e-05,
"loss": 3.783,
"step": 44500
},
{
"epoch": 0.17,
"learning_rate": 4.160297366692542e-05,
"loss": 3.7811,
"step": 45000
},
{
"epoch": 0.17,
"learning_rate": 4.150967337433571e-05,
"loss": 3.7804,
"step": 45500
},
{
"epoch": 0.17,
"learning_rate": 4.141637308174598e-05,
"loss": 3.7765,
"step": 46000
},
{
"epoch": 0.17,
"learning_rate": 4.132307278915627e-05,
"loss": 3.7656,
"step": 46500
},
{
"epoch": 0.18,
"learning_rate": 4.1229772496566546e-05,
"loss": 3.7627,
"step": 47000
},
{
"epoch": 0.18,
"learning_rate": 4.1136472203976835e-05,
"loss": 3.7595,
"step": 47500
},
{
"epoch": 0.18,
"learning_rate": 4.104317191138712e-05,
"loss": 3.7624,
"step": 48000
},
{
"epoch": 0.18,
"learning_rate": 4.09498716187974e-05,
"loss": 3.7571,
"step": 48500
},
{
"epoch": 0.18,
"learning_rate": 4.085657132620768e-05,
"loss": 3.7583,
"step": 49000
},
{
"epoch": 0.18,
"learning_rate": 4.076327103361796e-05,
"loss": 3.7583,
"step": 49500
},
{
"epoch": 0.19,
"learning_rate": 4.0669970741028244e-05,
"loss": 3.7491,
"step": 50000
},
{
"epoch": 0.19,
"eval_accuracy": 0.36325465564769804,
"eval_loss": 3.6769518852233887,
"eval_runtime": 6174.8828,
"eval_samples_per_second": 73.068,
"eval_steps_per_second": 1.142,
"step": 50000
},
{
"epoch": 0.19,
"learning_rate": 4.057667044843853e-05,
"loss": 3.7497,
"step": 50500
},
{
"epoch": 0.19,
"learning_rate": 4.048337015584881e-05,
"loss": 3.7401,
"step": 51000
},
{
"epoch": 0.19,
"learning_rate": 4.0390069863259097e-05,
"loss": 3.7423,
"step": 51500
},
{
"epoch": 0.19,
"learning_rate": 4.029676957066937e-05,
"loss": 3.7368,
"step": 52000
},
{
"epoch": 0.2,
"learning_rate": 4.020346927807966e-05,
"loss": 3.7358,
"step": 52500
},
{
"epoch": 0.2,
"learning_rate": 4.011016898548994e-05,
"loss": 3.7352,
"step": 53000
},
{
"epoch": 0.2,
"learning_rate": 4.0016868692900224e-05,
"loss": 3.7311,
"step": 53500
},
{
"epoch": 0.2,
"learning_rate": 3.9923568400310506e-05,
"loss": 3.7334,
"step": 54000
},
{
"epoch": 0.2,
"learning_rate": 3.983026810772079e-05,
"loss": 3.7347,
"step": 54500
},
{
"epoch": 0.21,
"learning_rate": 3.973696781513107e-05,
"loss": 3.7229,
"step": 55000
},
{
"epoch": 0.21,
"learning_rate": 3.964366752254135e-05,
"loss": 3.7265,
"step": 55500
},
{
"epoch": 0.21,
"learning_rate": 3.9550367229951633e-05,
"loss": 3.7286,
"step": 56000
},
{
"epoch": 0.21,
"learning_rate": 3.945706693736192e-05,
"loss": 3.7247,
"step": 56500
},
{
"epoch": 0.21,
"learning_rate": 3.93637666447722e-05,
"loss": 3.7186,
"step": 57000
},
{
"epoch": 0.21,
"learning_rate": 3.9270466352182486e-05,
"loss": 3.7152,
"step": 57500
},
{
"epoch": 0.22,
"learning_rate": 3.917716605959276e-05,
"loss": 3.713,
"step": 58000
},
{
"epoch": 0.22,
"learning_rate": 3.908386576700305e-05,
"loss": 3.7133,
"step": 58500
},
{
"epoch": 0.22,
"learning_rate": 3.899056547441333e-05,
"loss": 3.7076,
"step": 59000
},
{
"epoch": 0.22,
"learning_rate": 3.8897265181823613e-05,
"loss": 3.7082,
"step": 59500
},
{
"epoch": 0.22,
"learning_rate": 3.8803964889233895e-05,
"loss": 3.7062,
"step": 60000
},
{
"epoch": 0.22,
"eval_accuracy": 0.36794739205873755,
"eval_loss": 3.6288115978240967,
"eval_runtime": 6164.6808,
"eval_samples_per_second": 73.189,
"eval_steps_per_second": 1.144,
"step": 60000
},
{
"epoch": 0.23,
"learning_rate": 3.871066459664418e-05,
"loss": 3.7008,
"step": 60500
},
{
"epoch": 0.23,
"learning_rate": 3.861736430405446e-05,
"loss": 3.7013,
"step": 61000
},
{
"epoch": 0.23,
"learning_rate": 3.852406401146475e-05,
"loss": 3.6952,
"step": 61500
},
{
"epoch": 0.23,
"learning_rate": 3.843076371887502e-05,
"loss": 3.7065,
"step": 62000
},
{
"epoch": 0.23,
"learning_rate": 3.833746342628531e-05,
"loss": 3.692,
"step": 62500
},
{
"epoch": 0.24,
"learning_rate": 3.8244163133695587e-05,
"loss": 3.6942,
"step": 63000
},
{
"epoch": 0.24,
"learning_rate": 3.8150862841105875e-05,
"loss": 3.6906,
"step": 63500
},
{
"epoch": 0.24,
"learning_rate": 3.805756254851615e-05,
"loss": 3.6879,
"step": 64000
},
{
"epoch": 0.24,
"learning_rate": 3.796426225592644e-05,
"loss": 3.6885,
"step": 64500
},
{
"epoch": 0.24,
"learning_rate": 3.787096196333672e-05,
"loss": 3.6833,
"step": 65000
},
{
"epoch": 0.24,
"learning_rate": 3.7777661670747e-05,
"loss": 3.6834,
"step": 65500
},
{
"epoch": 0.25,
"learning_rate": 3.7684361378157285e-05,
"loss": 3.683,
"step": 66000
},
{
"epoch": 0.25,
"learning_rate": 3.7591061085567566e-05,
"loss": 3.6844,
"step": 66500
},
{
"epoch": 0.25,
"learning_rate": 3.749776079297785e-05,
"loss": 3.6726,
"step": 67000
},
{
"epoch": 0.25,
"learning_rate": 3.740446050038813e-05,
"loss": 3.6783,
"step": 67500
},
{
"epoch": 0.25,
"learning_rate": 3.731116020779841e-05,
"loss": 3.6768,
"step": 68000
},
{
"epoch": 0.26,
"learning_rate": 3.7217859915208694e-05,
"loss": 3.674,
"step": 68500
},
{
"epoch": 0.26,
"learning_rate": 3.7124559622618976e-05,
"loss": 3.6769,
"step": 69000
},
{
"epoch": 0.26,
"learning_rate": 3.703125933002926e-05,
"loss": 3.6695,
"step": 69500
},
{
"epoch": 0.26,
"learning_rate": 3.6937959037439546e-05,
"loss": 3.6724,
"step": 70000
},
{
"epoch": 0.26,
"eval_accuracy": 0.37135094159724996,
"eval_loss": 3.5937752723693848,
"eval_runtime": 6186.7336,
"eval_samples_per_second": 72.928,
"eval_steps_per_second": 1.14,
"step": 70000
},
{
"epoch": 0.26,
"learning_rate": 3.684465874484983e-05,
"loss": 3.6657,
"step": 70500
},
{
"epoch": 0.26,
"learning_rate": 3.675135845226011e-05,
"loss": 3.6618,
"step": 71000
},
{
"epoch": 0.27,
"learning_rate": 3.665805815967039e-05,
"loss": 3.6659,
"step": 71500
},
{
"epoch": 0.27,
"learning_rate": 3.6564757867080674e-05,
"loss": 3.6638,
"step": 72000
},
{
"epoch": 0.27,
"learning_rate": 3.6471457574490956e-05,
"loss": 3.6641,
"step": 72500
},
{
"epoch": 0.27,
"learning_rate": 3.637815728190124e-05,
"loss": 3.6576,
"step": 73000
},
{
"epoch": 0.27,
"learning_rate": 3.628485698931152e-05,
"loss": 3.6601,
"step": 73500
},
{
"epoch": 0.28,
"learning_rate": 3.61915566967218e-05,
"loss": 3.6542,
"step": 74000
},
{
"epoch": 0.28,
"learning_rate": 3.609825640413208e-05,
"loss": 3.6542,
"step": 74500
},
{
"epoch": 0.28,
"learning_rate": 3.6004956111542365e-05,
"loss": 3.658,
"step": 75000
},
{
"epoch": 0.28,
"learning_rate": 3.591165581895265e-05,
"loss": 3.6544,
"step": 75500
},
{
"epoch": 0.28,
"learning_rate": 3.5818355526362936e-05,
"loss": 3.6568,
"step": 76000
},
{
"epoch": 0.29,
"learning_rate": 3.572505523377321e-05,
"loss": 3.6491,
"step": 76500
},
{
"epoch": 0.29,
"learning_rate": 3.56317549411835e-05,
"loss": 3.6549,
"step": 77000
},
{
"epoch": 0.29,
"learning_rate": 3.5538454648593774e-05,
"loss": 3.6428,
"step": 77500
},
{
"epoch": 0.29,
"learning_rate": 3.544515435600406e-05,
"loss": 3.6446,
"step": 78000
},
{
"epoch": 0.29,
"learning_rate": 3.5351854063414345e-05,
"loss": 3.6421,
"step": 78500
},
{
"epoch": 0.29,
"learning_rate": 3.525855377082463e-05,
"loss": 3.6414,
"step": 79000
},
{
"epoch": 0.3,
"learning_rate": 3.516525347823491e-05,
"loss": 3.6386,
"step": 79500
},
{
"epoch": 0.3,
"learning_rate": 3.507195318564519e-05,
"loss": 3.6399,
"step": 80000
},
{
"epoch": 0.3,
"eval_accuracy": 0.37431053802265524,
"eval_loss": 3.5651655197143555,
"eval_runtime": 6163.5869,
"eval_samples_per_second": 73.202,
"eval_steps_per_second": 1.144,
"step": 80000
},
{
"epoch": 0.3,
"learning_rate": 3.497865289305547e-05,
"loss": 3.6369,
"step": 80500
},
{
"epoch": 0.3,
"learning_rate": 3.488535260046576e-05,
"loss": 3.6385,
"step": 81000
},
{
"epoch": 0.3,
"learning_rate": 3.4792052307876036e-05,
"loss": 3.6342,
"step": 81500
},
{
"epoch": 0.31,
"learning_rate": 3.4698752015286325e-05,
"loss": 3.6351,
"step": 82000
},
{
"epoch": 0.31,
"learning_rate": 3.46054517226966e-05,
"loss": 3.6348,
"step": 82500
},
{
"epoch": 0.31,
"learning_rate": 3.451215143010689e-05,
"loss": 3.6289,
"step": 83000
},
{
"epoch": 0.31,
"learning_rate": 3.4418851137517164e-05,
"loss": 3.6314,
"step": 83500
},
{
"epoch": 0.31,
"learning_rate": 3.432555084492745e-05,
"loss": 3.6277,
"step": 84000
},
{
"epoch": 0.32,
"learning_rate": 3.4232250552337734e-05,
"loss": 3.6296,
"step": 84500
},
{
"epoch": 0.32,
"learning_rate": 3.4138950259748016e-05,
"loss": 3.6274,
"step": 85000
},
{
"epoch": 0.32,
"learning_rate": 3.40456499671583e-05,
"loss": 3.62,
"step": 85500
},
{
"epoch": 0.32,
"learning_rate": 3.395234967456858e-05,
"loss": 3.6217,
"step": 86000
},
{
"epoch": 0.32,
"learning_rate": 3.385904938197886e-05,
"loss": 3.6205,
"step": 86500
},
{
"epoch": 0.32,
"learning_rate": 3.376574908938915e-05,
"loss": 3.6182,
"step": 87000
},
{
"epoch": 0.33,
"learning_rate": 3.3672448796799426e-05,
"loss": 3.6196,
"step": 87500
},
{
"epoch": 0.33,
"learning_rate": 3.3579148504209714e-05,
"loss": 3.6154,
"step": 88000
},
{
"epoch": 0.33,
"learning_rate": 3.348584821161999e-05,
"loss": 3.6242,
"step": 88500
},
{
"epoch": 0.33,
"learning_rate": 3.339254791903028e-05,
"loss": 3.6162,
"step": 89000
},
{
"epoch": 0.33,
"learning_rate": 3.329924762644056e-05,
"loss": 3.6205,
"step": 89500
},
{
"epoch": 0.34,
"learning_rate": 3.320594733385084e-05,
"loss": 3.6147,
"step": 90000
},
{
"epoch": 0.34,
"eval_accuracy": 0.3768022387359209,
"eval_loss": 3.539621591567993,
"eval_runtime": 6175.1379,
"eval_samples_per_second": 73.065,
"eval_steps_per_second": 1.142,
"step": 90000
},
{
"epoch": 0.34,
"learning_rate": 3.3112647041261124e-05,
"loss": 3.6185,
"step": 90500
},
{
"epoch": 0.34,
"learning_rate": 3.3019346748671405e-05,
"loss": 3.6125,
"step": 91000
},
{
"epoch": 0.34,
"learning_rate": 3.292604645608169e-05,
"loss": 3.6112,
"step": 91500
},
{
"epoch": 0.34,
"learning_rate": 3.283274616349197e-05,
"loss": 3.618,
"step": 92000
},
{
"epoch": 0.35,
"learning_rate": 3.273944587090225e-05,
"loss": 3.6135,
"step": 92500
},
{
"epoch": 0.35,
"learning_rate": 3.264614557831254e-05,
"loss": 3.6084,
"step": 93000
},
{
"epoch": 0.35,
"learning_rate": 3.2552845285722815e-05,
"loss": 3.606,
"step": 93500
},
{
"epoch": 0.35,
"learning_rate": 3.2459544993133103e-05,
"loss": 3.6125,
"step": 94000
},
{
"epoch": 0.35,
"learning_rate": 3.236624470054338e-05,
"loss": 3.6065,
"step": 94500
},
{
"epoch": 0.35,
"learning_rate": 3.227294440795367e-05,
"loss": 3.6035,
"step": 95000
},
{
"epoch": 0.36,
"learning_rate": 3.217964411536395e-05,
"loss": 3.5971,
"step": 95500
},
{
"epoch": 0.36,
"learning_rate": 3.208634382277423e-05,
"loss": 3.6011,
"step": 96000
},
{
"epoch": 0.36,
"learning_rate": 3.199304353018451e-05,
"loss": 3.601,
"step": 96500
},
{
"epoch": 0.36,
"learning_rate": 3.1899743237594795e-05,
"loss": 3.5967,
"step": 97000
},
{
"epoch": 0.36,
"learning_rate": 3.180644294500508e-05,
"loss": 3.6038,
"step": 97500
},
{
"epoch": 0.37,
"learning_rate": 3.171314265241536e-05,
"loss": 3.599,
"step": 98000
},
{
"epoch": 0.37,
"learning_rate": 3.161984235982564e-05,
"loss": 3.593,
"step": 98500
},
{
"epoch": 0.37,
"learning_rate": 3.152654206723593e-05,
"loss": 3.6019,
"step": 99000
},
{
"epoch": 0.37,
"learning_rate": 3.1433241774646204e-05,
"loss": 3.5927,
"step": 99500
},
{
"epoch": 0.37,
"learning_rate": 3.133994148205649e-05,
"loss": 3.5946,
"step": 100000
},
{
"epoch": 0.37,
"eval_accuracy": 0.3791379626568468,
"eval_loss": 3.515845775604248,
"eval_runtime": 6163.9176,
"eval_samples_per_second": 73.198,
"eval_steps_per_second": 1.144,
"step": 100000
},
{
"epoch": 0.38,
"learning_rate": 3.1246641189466775e-05,
"loss": 3.5983,
"step": 100500
},
{
"epoch": 0.38,
"learning_rate": 3.1153340896877057e-05,
"loss": 3.5947,
"step": 101000
},
{
"epoch": 0.38,
"learning_rate": 3.106004060428734e-05,
"loss": 3.5877,
"step": 101500
},
{
"epoch": 0.38,
"learning_rate": 3.096674031169762e-05,
"loss": 3.5896,
"step": 102000
},
{
"epoch": 0.38,
"learning_rate": 3.08734400191079e-05,
"loss": 3.5859,
"step": 102500
},
{
"epoch": 0.38,
"learning_rate": 3.0780139726518184e-05,
"loss": 3.5884,
"step": 103000
},
{
"epoch": 0.39,
"learning_rate": 3.0686839433928466e-05,
"loss": 3.5824,
"step": 103500
},
{
"epoch": 0.39,
"learning_rate": 3.059353914133875e-05,
"loss": 3.582,
"step": 104000
},
{
"epoch": 0.39,
"learning_rate": 3.050023884874903e-05,
"loss": 3.5859,
"step": 104500
},
{
"epoch": 0.39,
"learning_rate": 3.0406938556159315e-05,
"loss": 3.5828,
"step": 105000
},
{
"epoch": 0.39,
"learning_rate": 3.0313638263569593e-05,
"loss": 3.5827,
"step": 105500
},
{
"epoch": 0.4,
"learning_rate": 3.022033797097988e-05,
"loss": 3.58,
"step": 106000
},
{
"epoch": 0.4,
"learning_rate": 3.0127037678390164e-05,
"loss": 3.5793,
"step": 106500
},
{
"epoch": 0.4,
"learning_rate": 3.0033737385800442e-05,
"loss": 3.5823,
"step": 107000
},
{
"epoch": 0.4,
"learning_rate": 2.9940437093210728e-05,
"loss": 3.584,
"step": 107500
},
{
"epoch": 0.4,
"learning_rate": 2.9847136800621006e-05,
"loss": 3.5731,
"step": 108000
},
{
"epoch": 0.4,
"learning_rate": 2.975383650803129e-05,
"loss": 3.5804,
"step": 108500
},
{
"epoch": 0.41,
"learning_rate": 2.9660536215441577e-05,
"loss": 3.5809,
"step": 109000
},
{
"epoch": 0.41,
"learning_rate": 2.9567235922851855e-05,
"loss": 3.5782,
"step": 109500
},
{
"epoch": 0.41,
"learning_rate": 2.947393563026214e-05,
"loss": 3.5726,
"step": 110000
},
{
"epoch": 0.41,
"eval_accuracy": 0.3808981412485733,
"eval_loss": 3.4986438751220703,
"eval_runtime": 6162.0171,
"eval_samples_per_second": 73.22,
"eval_steps_per_second": 1.144,
"step": 110000
},
{
"epoch": 0.41,
"learning_rate": 2.938063533767242e-05,
"loss": 3.571,
"step": 110500
},
{
"epoch": 0.41,
"learning_rate": 2.9287335045082704e-05,
"loss": 3.5747,
"step": 111000
},
{
"epoch": 0.42,
"learning_rate": 2.9194034752492983e-05,
"loss": 3.5711,
"step": 111500
},
{
"epoch": 0.42,
"learning_rate": 2.9100734459903268e-05,
"loss": 3.574,
"step": 112000
},
{
"epoch": 0.42,
"learning_rate": 2.9007434167313553e-05,
"loss": 3.5739,
"step": 112500
},
{
"epoch": 0.42,
"learning_rate": 2.8914133874723832e-05,
"loss": 3.5682,
"step": 113000
},
{
"epoch": 0.42,
"learning_rate": 2.8820833582134117e-05,
"loss": 3.5678,
"step": 113500
},
{
"epoch": 0.43,
"learning_rate": 2.8727533289544395e-05,
"loss": 3.5692,
"step": 114000
},
{
"epoch": 0.43,
"learning_rate": 2.863423299695468e-05,
"loss": 3.5606,
"step": 114500
},
{
"epoch": 0.43,
"learning_rate": 2.8540932704364966e-05,
"loss": 3.5713,
"step": 115000
},
{
"epoch": 0.43,
"learning_rate": 2.8447632411775244e-05,
"loss": 3.5642,
"step": 115500
},
{
"epoch": 0.43,
"learning_rate": 2.835433211918553e-05,
"loss": 3.5661,
"step": 116000
},
{
"epoch": 0.43,
"learning_rate": 2.8261031826595808e-05,
"loss": 3.5661,
"step": 116500
},
{
"epoch": 0.44,
"learning_rate": 2.8167731534006093e-05,
"loss": 3.5653,
"step": 117000
},
{
"epoch": 0.44,
"learning_rate": 2.8074431241416375e-05,
"loss": 3.5709,
"step": 117500
},
{
"epoch": 0.44,
"learning_rate": 2.7981130948826657e-05,
"loss": 3.5604,
"step": 118000
},
{
"epoch": 0.44,
"learning_rate": 2.788783065623694e-05,
"loss": 3.5622,
"step": 118500
},
{
"epoch": 0.44,
"learning_rate": 2.779453036364722e-05,
"loss": 3.5606,
"step": 119000
},
{
"epoch": 0.45,
"learning_rate": 2.7701230071057503e-05,
"loss": 3.5591,
"step": 119500
},
{
"epoch": 0.45,
"learning_rate": 2.7607929778467788e-05,
"loss": 3.5631,
"step": 120000
},
{
"epoch": 0.45,
"eval_accuracy": 0.38260092346589303,
"eval_loss": 3.4819467067718506,
"eval_runtime": 6173.7982,
"eval_samples_per_second": 73.081,
"eval_steps_per_second": 1.142,
"step": 120000
},
{
"epoch": 0.45,
"learning_rate": 2.751462948587807e-05,
"loss": 3.5558,
"step": 120500
},
{
"epoch": 0.45,
"learning_rate": 2.7421329193288352e-05,
"loss": 3.5569,
"step": 121000
},
{
"epoch": 0.45,
"learning_rate": 2.7328028900698634e-05,
"loss": 3.559,
"step": 121500
},
{
"epoch": 0.46,
"learning_rate": 2.7234728608108916e-05,
"loss": 3.5596,
"step": 122000
},
{
"epoch": 0.46,
"learning_rate": 2.7141428315519198e-05,
"loss": 3.5582,
"step": 122500
},
{
"epoch": 0.46,
"learning_rate": 2.704812802292948e-05,
"loss": 3.553,
"step": 123000
},
{
"epoch": 0.46,
"learning_rate": 2.6954827730339765e-05,
"loss": 3.5566,
"step": 123500
},
{
"epoch": 0.46,
"learning_rate": 2.6861527437750043e-05,
"loss": 3.555,
"step": 124000
},
{
"epoch": 0.46,
"learning_rate": 2.676822714516033e-05,
"loss": 3.551,
"step": 124500
},
{
"epoch": 0.47,
"learning_rate": 2.667492685257061e-05,
"loss": 3.5519,
"step": 125000
},
{
"epoch": 0.47,
"learning_rate": 2.6581626559980892e-05,
"loss": 3.5491,
"step": 125500
},
{
"epoch": 0.47,
"learning_rate": 2.6488326267391177e-05,
"loss": 3.5511,
"step": 126000
},
{
"epoch": 0.47,
"learning_rate": 2.6395025974801456e-05,
"loss": 3.5442,
"step": 126500
},
{
"epoch": 0.47,
"learning_rate": 2.630172568221174e-05,
"loss": 3.5535,
"step": 127000
},
{
"epoch": 0.48,
"learning_rate": 2.620842538962202e-05,
"loss": 3.5475,
"step": 127500
},
{
"epoch": 0.48,
"learning_rate": 2.6115125097032305e-05,
"loss": 3.5421,
"step": 128000
},
{
"epoch": 0.48,
"learning_rate": 2.602182480444259e-05,
"loss": 3.5486,
"step": 128500
},
{
"epoch": 0.48,
"learning_rate": 2.592852451185287e-05,
"loss": 3.546,
"step": 129000
},
{
"epoch": 0.48,
"learning_rate": 2.5835224219263154e-05,
"loss": 3.5397,
"step": 129500
},
{
"epoch": 0.49,
"learning_rate": 2.5741923926673432e-05,
"loss": 3.5459,
"step": 130000
},
{
"epoch": 0.49,
"eval_accuracy": 0.384199481389569,
"eval_loss": 3.467806816101074,
"eval_runtime": 6189.2425,
"eval_samples_per_second": 72.898,
"eval_steps_per_second": 1.139,
"step": 130000
},
{
"epoch": 0.49,
"learning_rate": 2.5648623634083718e-05,
"loss": 3.5437,
"step": 130500
},
{
"epoch": 0.49,
"learning_rate": 2.5555323341493996e-05,
"loss": 3.5432,
"step": 131000
},
{
"epoch": 0.49,
"learning_rate": 2.546202304890428e-05,
"loss": 3.5416,
"step": 131500
},
{
"epoch": 0.49,
"learning_rate": 2.5368722756314567e-05,
"loss": 3.5455,
"step": 132000
},
{
"epoch": 0.49,
"learning_rate": 2.5275422463724845e-05,
"loss": 3.5381,
"step": 132500
},
{
"epoch": 0.5,
"learning_rate": 2.518212217113513e-05,
"loss": 3.541,
"step": 133000
},
{
"epoch": 0.5,
"learning_rate": 2.508882187854541e-05,
"loss": 3.5375,
"step": 133500
},
{
"epoch": 0.5,
"learning_rate": 2.4995521585955694e-05,
"loss": 3.5389,
"step": 134000
},
{
"epoch": 0.5,
"learning_rate": 2.4902221293365976e-05,
"loss": 3.5375,
"step": 134500
},
{
"epoch": 0.5,
"learning_rate": 2.4808921000776258e-05,
"loss": 3.5344,
"step": 135000
},
{
"epoch": 0.51,
"learning_rate": 2.4715620708186543e-05,
"loss": 3.5342,
"step": 135500
},
{
"epoch": 0.51,
"learning_rate": 2.4622320415596825e-05,
"loss": 3.5444,
"step": 136000
},
{
"epoch": 0.51,
"learning_rate": 2.4529020123007107e-05,
"loss": 3.5318,
"step": 136500
},
{
"epoch": 0.51,
"learning_rate": 2.443571983041739e-05,
"loss": 3.5402,
"step": 137000
},
{
"epoch": 0.51,
"learning_rate": 2.434241953782767e-05,
"loss": 3.5408,
"step": 137500
},
{
"epoch": 0.52,
"learning_rate": 2.4249119245237953e-05,
"loss": 3.5314,
"step": 138000
},
{
"epoch": 0.52,
"learning_rate": 2.4155818952648238e-05,
"loss": 3.5276,
"step": 138500
},
{
"epoch": 0.52,
"learning_rate": 2.406251866005852e-05,
"loss": 3.5309,
"step": 139000
},
{
"epoch": 0.52,
"learning_rate": 2.39692183674688e-05,
"loss": 3.533,
"step": 139500
},
{
"epoch": 0.52,
"learning_rate": 2.3875918074879084e-05,
"loss": 3.5304,
"step": 140000
},
{
"epoch": 0.52,
"eval_accuracy": 0.3857324143630419,
"eval_loss": 3.4535393714904785,
"eval_runtime": 6160.8797,
"eval_samples_per_second": 73.234,
"eval_steps_per_second": 1.144,
"step": 140000
},
{
"epoch": 0.52,
"learning_rate": 2.3782617782289365e-05,
"loss": 3.536,
"step": 140500
},
{
"epoch": 0.53,
"learning_rate": 2.368931748969965e-05,
"loss": 3.5374,
"step": 141000
},
{
"epoch": 0.53,
"learning_rate": 2.3596017197109933e-05,
"loss": 3.5279,
"step": 141500
},
{
"epoch": 0.53,
"learning_rate": 2.3502716904520214e-05,
"loss": 3.5334,
"step": 142000
},
{
"epoch": 0.53,
"learning_rate": 2.3409416611930496e-05,
"loss": 3.531,
"step": 142500
},
{
"epoch": 0.53,
"learning_rate": 2.3316116319340778e-05,
"loss": 3.5285,
"step": 143000
},
{
"epoch": 0.54,
"learning_rate": 2.322281602675106e-05,
"loss": 3.5266,
"step": 143500
},
{
"epoch": 0.54,
"learning_rate": 2.3129515734161345e-05,
"loss": 3.5307,
"step": 144000
},
{
"epoch": 0.54,
"learning_rate": 2.3036215441571627e-05,
"loss": 3.534,
"step": 144500
},
{
"epoch": 0.54,
"learning_rate": 2.294291514898191e-05,
"loss": 3.5278,
"step": 145000
},
{
"epoch": 0.54,
"learning_rate": 2.284961485639219e-05,
"loss": 3.522,
"step": 145500
},
{
"epoch": 0.54,
"learning_rate": 2.2756314563802473e-05,
"loss": 3.5234,
"step": 146000
},
{
"epoch": 0.55,
"learning_rate": 2.2663014271212758e-05,
"loss": 3.5258,
"step": 146500
},
{
"epoch": 0.55,
"learning_rate": 2.256971397862304e-05,
"loss": 3.5226,
"step": 147000
},
{
"epoch": 0.55,
"learning_rate": 2.2476413686033322e-05,
"loss": 3.5204,
"step": 147500
},
{
"epoch": 0.55,
"learning_rate": 2.2383113393443604e-05,
"loss": 3.5237,
"step": 148000
},
{
"epoch": 0.55,
"learning_rate": 2.2289813100853886e-05,
"loss": 3.5225,
"step": 148500
},
{
"epoch": 0.56,
"learning_rate": 2.2196512808264167e-05,
"loss": 3.5218,
"step": 149000
},
{
"epoch": 0.56,
"learning_rate": 2.2103212515674453e-05,
"loss": 3.5269,
"step": 149500
},
{
"epoch": 0.56,
"learning_rate": 2.2009912223084735e-05,
"loss": 3.5245,
"step": 150000
},
{
"epoch": 0.56,
"eval_accuracy": 0.3867262152967859,
"eval_loss": 3.442957639694214,
"eval_runtime": 6174.4314,
"eval_samples_per_second": 73.073,
"eval_steps_per_second": 1.142,
"step": 150000
},
{
"epoch": 0.56,
"learning_rate": 2.1916611930495016e-05,
"loss": 3.5227,
"step": 150500
},
{
"epoch": 0.56,
"learning_rate": 2.18233116379053e-05,
"loss": 3.5117,
"step": 151000
},
{
"epoch": 0.57,
"learning_rate": 2.173001134531558e-05,
"loss": 3.5203,
"step": 151500
},
{
"epoch": 0.57,
"learning_rate": 2.1636711052725862e-05,
"loss": 3.5177,
"step": 152000
},
{
"epoch": 0.57,
"learning_rate": 2.1543410760136147e-05,
"loss": 3.5175,
"step": 152500
},
{
"epoch": 0.57,
"learning_rate": 2.145011046754643e-05,
"loss": 3.5186,
"step": 153000
},
{
"epoch": 0.57,
"learning_rate": 2.135681017495671e-05,
"loss": 3.5208,
"step": 153500
},
{
"epoch": 0.57,
"learning_rate": 2.1263509882366993e-05,
"loss": 3.5199,
"step": 154000
},
{
"epoch": 0.58,
"learning_rate": 2.1170209589777275e-05,
"loss": 3.5188,
"step": 154500
},
{
"epoch": 0.58,
"learning_rate": 2.1076909297187557e-05,
"loss": 3.5235,
"step": 155000
},
{
"epoch": 0.58,
"learning_rate": 2.098360900459784e-05,
"loss": 3.516,
"step": 155500
},
{
"epoch": 0.58,
"learning_rate": 2.089030871200812e-05,
"loss": 3.5124,
"step": 156000
},
{
"epoch": 0.58,
"learning_rate": 2.0797008419418402e-05,
"loss": 3.5145,
"step": 156500
},
{
"epoch": 0.59,
"learning_rate": 2.0703708126828684e-05,
"loss": 3.5092,
"step": 157000
},
{
"epoch": 0.59,
"learning_rate": 2.061040783423897e-05,
"loss": 3.5142,
"step": 157500
},
{
"epoch": 0.59,
"learning_rate": 2.051710754164925e-05,
"loss": 3.5084,
"step": 158000
},
{
"epoch": 0.59,
"learning_rate": 2.0423807249059533e-05,
"loss": 3.5147,
"step": 158500
},
{
"epoch": 0.59,
"learning_rate": 2.0330506956469815e-05,
"loss": 3.5112,
"step": 159000
},
{
"epoch": 0.6,
"learning_rate": 2.0237206663880097e-05,
"loss": 3.5137,
"step": 159500
},
{
"epoch": 0.6,
"learning_rate": 2.014390637129038e-05,
"loss": 3.5124,
"step": 160000
},
{
"epoch": 0.6,
"eval_accuracy": 0.3877192016058592,
"eval_loss": 3.432858943939209,
"eval_runtime": 6161.9947,
"eval_samples_per_second": 73.221,
"eval_steps_per_second": 1.144,
"step": 160000
},
{
"epoch": 0.6,
"learning_rate": 2.0050606078700664e-05,
"loss": 3.5108,
"step": 160500
},
{
"epoch": 0.6,
"learning_rate": 1.9957305786110946e-05,
"loss": 3.5035,
"step": 161000
},
{
"epoch": 0.6,
"learning_rate": 1.9864005493521228e-05,
"loss": 3.5044,
"step": 161500
},
{
"epoch": 0.6,
"learning_rate": 1.977070520093151e-05,
"loss": 3.5069,
"step": 162000
},
{
"epoch": 0.61,
"learning_rate": 1.967740490834179e-05,
"loss": 3.5078,
"step": 162500
},
{
"epoch": 0.61,
"learning_rate": 1.9584104615752074e-05,
"loss": 3.5003,
"step": 163000
},
{
"epoch": 0.61,
"learning_rate": 1.949080432316236e-05,
"loss": 3.5056,
"step": 163500
},
{
"epoch": 0.61,
"learning_rate": 1.939750403057264e-05,
"loss": 3.5082,
"step": 164000
},
{
"epoch": 0.61,
"learning_rate": 1.9304203737982923e-05,
"loss": 3.5086,
"step": 164500
},
{
"epoch": 0.62,
"learning_rate": 1.9210903445393204e-05,
"loss": 3.4997,
"step": 165000
},
{
"epoch": 0.62,
"learning_rate": 1.9117603152803486e-05,
"loss": 3.5013,
"step": 165500
},
{
"epoch": 0.62,
"learning_rate": 1.902430286021377e-05,
"loss": 3.5048,
"step": 166000
},
{
"epoch": 0.62,
"learning_rate": 1.8931002567624053e-05,
"loss": 3.5067,
"step": 166500
},
{
"epoch": 0.62,
"learning_rate": 1.8837702275034335e-05,
"loss": 3.5029,
"step": 167000
},
{
"epoch": 0.63,
"learning_rate": 1.8744401982444617e-05,
"loss": 3.503,
"step": 167500
},
{
"epoch": 0.63,
"learning_rate": 1.86511016898549e-05,
"loss": 3.5041,
"step": 168000
},
{
"epoch": 0.63,
"learning_rate": 1.855780139726518e-05,
"loss": 3.5038,
"step": 168500
},
{
"epoch": 0.63,
"learning_rate": 1.8464501104675466e-05,
"loss": 3.5015,
"step": 169000
},
{
"epoch": 0.63,
"learning_rate": 1.8371200812085748e-05,
"loss": 3.4985,
"step": 169500
},
{
"epoch": 0.63,
"learning_rate": 1.827790051949603e-05,
"loss": 3.501,
"step": 170000
},
{
"epoch": 0.63,
"eval_accuracy": 0.3889908155509813,
"eval_loss": 3.422307252883911,
"eval_runtime": 6181.3169,
"eval_samples_per_second": 72.992,
"eval_steps_per_second": 1.141,
"step": 170000
},
{
"epoch": 0.64,
"learning_rate": 1.8184600226906312e-05,
"loss": 3.4997,
"step": 170500
},
{
"epoch": 0.64,
"learning_rate": 1.8091299934316594e-05,
"loss": 3.5017,
"step": 171000
},
{
"epoch": 0.64,
"learning_rate": 1.799799964172688e-05,
"loss": 3.4982,
"step": 171500
},
{
"epoch": 0.64,
"learning_rate": 1.790469934913716e-05,
"loss": 3.4967,
"step": 172000
},
{
"epoch": 0.64,
"learning_rate": 1.7811399056547443e-05,
"loss": 3.4937,
"step": 172500
},
{
"epoch": 0.65,
"learning_rate": 1.7718098763957725e-05,
"loss": 3.5034,
"step": 173000
},
{
"epoch": 0.65,
"learning_rate": 1.7624798471368006e-05,
"loss": 3.494,
"step": 173500
},
{
"epoch": 0.65,
"learning_rate": 1.753149817877829e-05,
"loss": 3.5028,
"step": 174000
},
{
"epoch": 0.65,
"learning_rate": 1.7438197886188574e-05,
"loss": 3.4946,
"step": 174500
},
{
"epoch": 0.65,
"learning_rate": 1.7344897593598855e-05,
"loss": 3.4916,
"step": 175000
},
{
"epoch": 0.65,
"learning_rate": 1.7251597301009137e-05,
"loss": 3.4945,
"step": 175500
},
{
"epoch": 0.66,
"learning_rate": 1.715829700841942e-05,
"loss": 3.4958,
"step": 176000
},
{
"epoch": 0.66,
"learning_rate": 1.70649967158297e-05,
"loss": 3.4939,
"step": 176500
},
{
"epoch": 0.66,
"learning_rate": 1.6971696423239983e-05,
"loss": 3.4976,
"step": 177000
},
{
"epoch": 0.66,
"learning_rate": 1.6878396130650268e-05,
"loss": 3.4994,
"step": 177500
},
{
"epoch": 0.66,
"learning_rate": 1.678509583806055e-05,
"loss": 3.4868,
"step": 178000
},
{
"epoch": 0.67,
"learning_rate": 1.6691795545470832e-05,
"loss": 3.4933,
"step": 178500
},
{
"epoch": 0.67,
"learning_rate": 1.6598495252881114e-05,
"loss": 3.4922,
"step": 179000
},
{
"epoch": 0.67,
"learning_rate": 1.6505194960291396e-05,
"loss": 3.4896,
"step": 179500
},
{
"epoch": 0.67,
"learning_rate": 1.641189466770168e-05,
"loss": 3.4934,
"step": 180000
},
{
"epoch": 0.67,
"eval_accuracy": 0.3900964280538928,
"eval_loss": 3.4130406379699707,
"eval_runtime": 6165.5394,
"eval_samples_per_second": 73.179,
"eval_steps_per_second": 1.143,
"step": 180000
},
{
"epoch": 0.67,
"learning_rate": 1.6318594375111963e-05,
"loss": 3.4938,
"step": 180500
},
{
"epoch": 0.68,
"learning_rate": 1.6225294082522245e-05,
"loss": 3.4904,
"step": 181000
},
{
"epoch": 0.68,
"learning_rate": 1.6131993789932527e-05,
"loss": 3.4894,
"step": 181500
},
{
"epoch": 0.68,
"learning_rate": 1.603869349734281e-05,
"loss": 3.49,
"step": 182000
},
{
"epoch": 0.68,
"learning_rate": 1.594539320475309e-05,
"loss": 3.4879,
"step": 182500
},
{
"epoch": 0.68,
"learning_rate": 1.5852092912163376e-05,
"loss": 3.4914,
"step": 183000
},
{
"epoch": 0.68,
"learning_rate": 1.5758792619573658e-05,
"loss": 3.4971,
"step": 183500
},
{
"epoch": 0.69,
"learning_rate": 1.566549232698394e-05,
"loss": 3.4892,
"step": 184000
},
{
"epoch": 0.69,
"learning_rate": 1.557219203439422e-05,
"loss": 3.4822,
"step": 184500
},
{
"epoch": 0.69,
"learning_rate": 1.5478891741804503e-05,
"loss": 3.4924,
"step": 185000
},
{
"epoch": 0.69,
"learning_rate": 1.538559144921479e-05,
"loss": 3.4869,
"step": 185500
},
{
"epoch": 0.69,
"learning_rate": 1.529229115662507e-05,
"loss": 3.4916,
"step": 186000
},
{
"epoch": 0.7,
"learning_rate": 1.519899086403535e-05,
"loss": 3.4782,
"step": 186500
},
{
"epoch": 0.7,
"learning_rate": 1.5105690571445632e-05,
"loss": 3.4797,
"step": 187000
},
{
"epoch": 0.7,
"learning_rate": 1.5012390278855914e-05,
"loss": 3.4762,
"step": 187500
},
{
"epoch": 0.7,
"learning_rate": 1.4919089986266196e-05,
"loss": 3.4915,
"step": 188000
},
{
"epoch": 0.7,
"learning_rate": 1.4825789693676481e-05,
"loss": 3.4816,
"step": 188500
},
{
"epoch": 0.71,
"learning_rate": 1.4732489401086763e-05,
"loss": 3.4886,
"step": 189000
},
{
"epoch": 0.71,
"learning_rate": 1.4639189108497045e-05,
"loss": 3.4776,
"step": 189500
},
{
"epoch": 0.71,
"learning_rate": 1.4545888815907327e-05,
"loss": 3.4863,
"step": 190000
},
{
"epoch": 0.71,
"eval_accuracy": 0.390879254197248,
"eval_loss": 3.4041781425476074,
"eval_runtime": 6176.7947,
"eval_samples_per_second": 73.045,
"eval_steps_per_second": 1.141,
"step": 190000
},
{
"epoch": 0.71,
"learning_rate": 1.4452588523317609e-05,
"loss": 3.4775,
"step": 190500
},
{
"epoch": 0.71,
"learning_rate": 1.435928823072789e-05,
"loss": 3.4816,
"step": 191000
},
{
"epoch": 0.71,
"learning_rate": 1.4265987938138176e-05,
"loss": 3.4877,
"step": 191500
},
{
"epoch": 0.72,
"learning_rate": 1.4172687645548458e-05,
"loss": 3.479,
"step": 192000
},
{
"epoch": 0.72,
"learning_rate": 1.407938735295874e-05,
"loss": 3.4829,
"step": 192500
},
{
"epoch": 0.72,
"learning_rate": 1.3986087060369022e-05,
"loss": 3.4774,
"step": 193000
},
{
"epoch": 0.72,
"learning_rate": 1.3892786767779304e-05,
"loss": 3.4793,
"step": 193500
},
{
"epoch": 0.72,
"learning_rate": 1.3799486475189589e-05,
"loss": 3.4811,
"step": 194000
},
{
"epoch": 0.73,
"learning_rate": 1.370618618259987e-05,
"loss": 3.4737,
"step": 194500
},
{
"epoch": 0.73,
"learning_rate": 1.3612885890010153e-05,
"loss": 3.4784,
"step": 195000
},
{
"epoch": 0.73,
"learning_rate": 1.3519585597420434e-05,
"loss": 3.4769,
"step": 195500
},
{
"epoch": 0.73,
"learning_rate": 1.3426285304830716e-05,
"loss": 3.4766,
"step": 196000
},
{
"epoch": 0.73,
"learning_rate": 1.3332985012240998e-05,
"loss": 3.48,
"step": 196500
},
{
"epoch": 0.74,
"learning_rate": 1.3239684719651282e-05,
"loss": 3.4781,
"step": 197000
},
{
"epoch": 0.74,
"learning_rate": 1.3146384427061564e-05,
"loss": 3.4774,
"step": 197500
},
{
"epoch": 0.74,
"learning_rate": 1.3053084134471845e-05,
"loss": 3.4706,
"step": 198000
},
{
"epoch": 0.74,
"learning_rate": 1.2959783841882129e-05,
"loss": 3.474,
"step": 198500
},
{
"epoch": 0.74,
"learning_rate": 1.2866483549292411e-05,
"loss": 3.4722,
"step": 199000
},
{
"epoch": 0.74,
"learning_rate": 1.2773183256702694e-05,
"loss": 3.4734,
"step": 199500
},
{
"epoch": 0.75,
"learning_rate": 1.2679882964112976e-05,
"loss": 3.4799,
"step": 200000
},
{
"epoch": 0.75,
"eval_accuracy": 0.3914298624786812,
"eval_loss": 3.39914870262146,
"eval_runtime": 6175.7078,
"eval_samples_per_second": 73.058,
"eval_steps_per_second": 1.142,
"step": 200000
},
{
"epoch": 0.75,
"learning_rate": 1.2586582671523258e-05,
"loss": 3.4762,
"step": 200500
},
{
"epoch": 0.75,
"learning_rate": 1.249328237893354e-05,
"loss": 3.4779,
"step": 201000
},
{
"epoch": 0.75,
"learning_rate": 1.2399982086343824e-05,
"loss": 3.476,
"step": 201500
},
{
"epoch": 0.75,
"learning_rate": 1.2306681793754106e-05,
"loss": 3.473,
"step": 202000
},
{
"epoch": 0.76,
"learning_rate": 1.2213381501164387e-05,
"loss": 3.4828,
"step": 202500
},
{
"epoch": 0.76,
"learning_rate": 1.2120081208574671e-05,
"loss": 3.4759,
"step": 203000
},
{
"epoch": 0.76,
"learning_rate": 1.2026780915984953e-05,
"loss": 3.4801,
"step": 203500
},
{
"epoch": 0.76,
"learning_rate": 1.1933480623395235e-05,
"loss": 3.4724,
"step": 204000
},
{
"epoch": 0.76,
"learning_rate": 1.1840180330805518e-05,
"loss": 3.4795,
"step": 204500
},
{
"epoch": 0.77,
"learning_rate": 1.17468800382158e-05,
"loss": 3.4742,
"step": 205000
},
{
"epoch": 0.77,
"learning_rate": 1.1653579745626082e-05,
"loss": 3.4712,
"step": 205500
},
{
"epoch": 0.77,
"learning_rate": 1.1560279453036366e-05,
"loss": 3.4743,
"step": 206000
},
{
"epoch": 0.77,
"learning_rate": 1.1466979160446648e-05,
"loss": 3.471,
"step": 206500
},
{
"epoch": 0.77,
"learning_rate": 1.137367886785693e-05,
"loss": 3.4698,
"step": 207000
},
{
"epoch": 0.77,
"learning_rate": 1.1280378575267213e-05,
"loss": 3.4809,
"step": 207500
},
{
"epoch": 0.78,
"learning_rate": 1.1187078282677495e-05,
"loss": 3.472,
"step": 208000
},
{
"epoch": 0.78,
"learning_rate": 1.1093777990087778e-05,
"loss": 3.4675,
"step": 208500
},
{
"epoch": 0.78,
"learning_rate": 1.100047769749806e-05,
"loss": 3.4722,
"step": 209000
},
{
"epoch": 0.78,
"learning_rate": 1.0907177404908342e-05,
"loss": 3.4684,
"step": 209500
},
{
"epoch": 0.78,
"learning_rate": 1.0813877112318626e-05,
"loss": 3.4682,
"step": 210000
},
{
"epoch": 0.78,
"eval_accuracy": 0.3923717267565564,
"eval_loss": 3.3909456729888916,
"eval_runtime": 6173.3331,
"eval_samples_per_second": 73.086,
"eval_steps_per_second": 1.142,
"step": 210000
},
{
"epoch": 0.79,
"learning_rate": 1.0720576819728908e-05,
"loss": 3.465,
"step": 210500
},
{
"epoch": 0.79,
"learning_rate": 1.062727652713919e-05,
"loss": 3.4655,
"step": 211000
},
{
"epoch": 0.79,
"learning_rate": 1.0533976234549473e-05,
"loss": 3.4601,
"step": 211500
},
{
"epoch": 0.79,
"learning_rate": 1.0440675941959755e-05,
"loss": 3.4688,
"step": 212000
},
{
"epoch": 0.79,
"learning_rate": 1.0347375649370037e-05,
"loss": 3.4694,
"step": 212500
},
{
"epoch": 0.79,
"learning_rate": 1.025407535678032e-05,
"loss": 3.4675,
"step": 213000
},
{
"epoch": 0.8,
"learning_rate": 1.0160775064190602e-05,
"loss": 3.4701,
"step": 213500
},
{
"epoch": 0.8,
"learning_rate": 1.0067474771600884e-05,
"loss": 3.4659,
"step": 214000
},
{
"epoch": 0.8,
"learning_rate": 9.974174479011166e-06,
"loss": 3.4678,
"step": 214500
},
{
"epoch": 0.8,
"learning_rate": 9.88087418642145e-06,
"loss": 3.4694,
"step": 215000
},
{
"epoch": 0.8,
"learning_rate": 9.787573893831731e-06,
"loss": 3.4714,
"step": 215500
},
{
"epoch": 0.81,
"learning_rate": 9.694273601242013e-06,
"loss": 3.4613,
"step": 216000
},
{
"epoch": 0.81,
"learning_rate": 9.600973308652295e-06,
"loss": 3.4647,
"step": 216500
},
{
"epoch": 0.81,
"learning_rate": 9.507673016062579e-06,
"loss": 3.4664,
"step": 217000
},
{
"epoch": 0.81,
"learning_rate": 9.41437272347286e-06,
"loss": 3.4631,
"step": 217500
},
{
"epoch": 0.81,
"learning_rate": 9.321072430883143e-06,
"loss": 3.4666,
"step": 218000
},
{
"epoch": 0.82,
"learning_rate": 9.227772138293426e-06,
"loss": 3.467,
"step": 218500
},
{
"epoch": 0.82,
"learning_rate": 9.134471845703708e-06,
"loss": 3.4636,
"step": 219000
},
{
"epoch": 0.82,
"learning_rate": 9.04117155311399e-06,
"loss": 3.4637,
"step": 219500
},
{
"epoch": 0.82,
"learning_rate": 8.947871260524273e-06,
"loss": 3.4667,
"step": 220000
},
{
"epoch": 0.82,
"eval_accuracy": 0.3929634584179766,
"eval_loss": 3.385192394256592,
"eval_runtime": 6176.1319,
"eval_samples_per_second": 73.053,
"eval_steps_per_second": 1.141,
"step": 220000
},
{
"epoch": 0.82,
"learning_rate": 8.854570967934555e-06,
"loss": 3.4613,
"step": 220500
},
{
"epoch": 0.82,
"learning_rate": 8.761270675344839e-06,
"loss": 3.4618,
"step": 221000
},
{
"epoch": 0.83,
"learning_rate": 8.66797038275512e-06,
"loss": 3.4675,
"step": 221500
},
{
"epoch": 0.83,
"learning_rate": 8.574670090165403e-06,
"loss": 3.4575,
"step": 222000
},
{
"epoch": 0.83,
"learning_rate": 8.481369797575686e-06,
"loss": 3.4609,
"step": 222500
},
{
"epoch": 0.83,
"learning_rate": 8.388069504985968e-06,
"loss": 3.4639,
"step": 223000
},
{
"epoch": 0.83,
"learning_rate": 8.29476921239625e-06,
"loss": 3.4629,
"step": 223500
},
{
"epoch": 0.84,
"learning_rate": 8.201468919806534e-06,
"loss": 3.4579,
"step": 224000
},
{
"epoch": 0.84,
"learning_rate": 8.108168627216815e-06,
"loss": 3.461,
"step": 224500
},
{
"epoch": 0.84,
"learning_rate": 8.014868334627097e-06,
"loss": 3.4586,
"step": 225000
},
{
"epoch": 0.84,
"learning_rate": 7.92156804203738e-06,
"loss": 3.4619,
"step": 225500
},
{
"epoch": 0.84,
"learning_rate": 7.828267749447663e-06,
"loss": 3.4657,
"step": 226000
},
{
"epoch": 0.85,
"learning_rate": 7.734967456857945e-06,
"loss": 3.4628,
"step": 226500
},
{
"epoch": 0.85,
"learning_rate": 7.641667164268228e-06,
"loss": 3.461,
"step": 227000
},
{
"epoch": 0.85,
"learning_rate": 7.54836687167851e-06,
"loss": 3.4642,
"step": 227500
},
{
"epoch": 0.85,
"learning_rate": 7.455066579088793e-06,
"loss": 3.459,
"step": 228000
},
{
"epoch": 0.85,
"learning_rate": 7.361766286499075e-06,
"loss": 3.4579,
"step": 228500
},
{
"epoch": 0.85,
"learning_rate": 7.2684659939093565e-06,
"loss": 3.4543,
"step": 229000
},
{
"epoch": 0.86,
"learning_rate": 7.17516570131964e-06,
"loss": 3.4529,
"step": 229500
},
{
"epoch": 0.86,
"learning_rate": 7.081865408729922e-06,
"loss": 3.4564,
"step": 230000
},
{
"epoch": 0.86,
"eval_accuracy": 0.39360805835390505,
"eval_loss": 3.3790297508239746,
"eval_runtime": 6181.6044,
"eval_samples_per_second": 72.988,
"eval_steps_per_second": 1.14,
"step": 230000
},
{
"epoch": 0.86,
"learning_rate": 6.988565116140204e-06,
"loss": 3.4576,
"step": 230500
},
{
"epoch": 0.86,
"learning_rate": 6.895264823550487e-06,
"loss": 3.4563,
"step": 231000
},
{
"epoch": 0.86,
"learning_rate": 6.801964530960769e-06,
"loss": 3.4598,
"step": 231500
},
{
"epoch": 0.87,
"learning_rate": 6.708664238371051e-06,
"loss": 3.4568,
"step": 232000
},
{
"epoch": 0.87,
"learning_rate": 6.615363945781335e-06,
"loss": 3.4592,
"step": 232500
},
{
"epoch": 0.87,
"learning_rate": 6.522063653191617e-06,
"loss": 3.4551,
"step": 233000
},
{
"epoch": 0.87,
"learning_rate": 6.4287633606018985e-06,
"loss": 3.4606,
"step": 233500
},
{
"epoch": 0.87,
"learning_rate": 6.335463068012182e-06,
"loss": 3.4538,
"step": 234000
},
{
"epoch": 0.88,
"learning_rate": 6.242162775422464e-06,
"loss": 3.4554,
"step": 234500
},
{
"epoch": 0.88,
"learning_rate": 6.148862482832747e-06,
"loss": 3.4551,
"step": 235000
},
{
"epoch": 0.88,
"learning_rate": 6.055562190243029e-06,
"loss": 3.4592,
"step": 235500
},
{
"epoch": 0.88,
"learning_rate": 5.962261897653311e-06,
"loss": 3.4535,
"step": 236000
},
{
"epoch": 0.88,
"learning_rate": 5.868961605063594e-06,
"loss": 3.4534,
"step": 236500
},
{
"epoch": 0.88,
"learning_rate": 5.775661312473876e-06,
"loss": 3.458,
"step": 237000
},
{
"epoch": 0.89,
"learning_rate": 5.6823610198841586e-06,
"loss": 3.4605,
"step": 237500
},
{
"epoch": 0.89,
"learning_rate": 5.5890607272944404e-06,
"loss": 3.4597,
"step": 238000
},
{
"epoch": 0.89,
"learning_rate": 5.495760434704723e-06,
"loss": 3.4539,
"step": 238500
},
{
"epoch": 0.89,
"learning_rate": 5.402460142115006e-06,
"loss": 3.4585,
"step": 239000
},
{
"epoch": 0.89,
"learning_rate": 5.309159849525289e-06,
"loss": 3.4474,
"step": 239500
},
{
"epoch": 0.9,
"learning_rate": 5.2158595569355705e-06,
"loss": 3.4581,
"step": 240000
},
{
"epoch": 0.9,
"eval_accuracy": 0.3941211310703905,
"eval_loss": 3.3752598762512207,
"eval_runtime": 6171.8723,
"eval_samples_per_second": 73.103,
"eval_steps_per_second": 1.142,
"step": 240000
},
{
"epoch": 0.9,
"learning_rate": 5.122559264345853e-06,
"loss": 3.4558,
"step": 240500
},
{
"epoch": 0.9,
"learning_rate": 5.029258971756136e-06,
"loss": 3.4538,
"step": 241000
},
{
"epoch": 0.9,
"learning_rate": 4.935958679166418e-06,
"loss": 3.4561,
"step": 241500
},
{
"epoch": 0.9,
"learning_rate": 4.8426583865767005e-06,
"loss": 3.4564,
"step": 242000
},
{
"epoch": 0.91,
"learning_rate": 4.749358093986983e-06,
"loss": 3.4544,
"step": 242500
},
{
"epoch": 0.91,
"learning_rate": 4.656057801397266e-06,
"loss": 3.4481,
"step": 243000
},
{
"epoch": 0.91,
"learning_rate": 4.562757508807548e-06,
"loss": 3.4535,
"step": 243500
},
{
"epoch": 0.91,
"learning_rate": 4.4694572162178306e-06,
"loss": 3.4538,
"step": 244000
},
{
"epoch": 0.91,
"learning_rate": 4.376156923628113e-06,
"loss": 3.4467,
"step": 244500
},
{
"epoch": 0.91,
"learning_rate": 4.282856631038395e-06,
"loss": 3.4495,
"step": 245000
},
{
"epoch": 0.92,
"learning_rate": 4.189556338448677e-06,
"loss": 3.4485,
"step": 245500
},
{
"epoch": 0.92,
"learning_rate": 4.09625604585896e-06,
"loss": 3.4535,
"step": 246000
},
{
"epoch": 0.92,
"learning_rate": 4.0029557532692425e-06,
"loss": 3.45,
"step": 246500
},
{
"epoch": 0.92,
"learning_rate": 3.909655460679524e-06,
"loss": 3.4558,
"step": 247000
},
{
"epoch": 0.92,
"learning_rate": 3.816355168089807e-06,
"loss": 3.4502,
"step": 247500
},
{
"epoch": 0.93,
"learning_rate": 3.72305487550009e-06,
"loss": 3.4509,
"step": 248000
},
{
"epoch": 0.93,
"learning_rate": 3.6297545829103717e-06,
"loss": 3.4575,
"step": 248500
},
{
"epoch": 0.93,
"learning_rate": 3.5364542903206544e-06,
"loss": 3.451,
"step": 249000
},
{
"epoch": 0.93,
"learning_rate": 3.443153997730937e-06,
"loss": 3.4456,
"step": 249500
},
{
"epoch": 0.93,
"learning_rate": 3.34985370514122e-06,
"loss": 3.4553,
"step": 250000
},
{
"epoch": 0.93,
"eval_accuracy": 0.3945191770501251,
"eval_loss": 3.370955467224121,
"eval_runtime": 6175.512,
"eval_samples_per_second": 73.06,
"eval_steps_per_second": 1.142,
"step": 250000
},
{
"epoch": 0.93,
"learning_rate": 3.2565534125515017e-06,
"loss": 3.4459,
"step": 250500
},
{
"epoch": 0.94,
"learning_rate": 3.1632531199617845e-06,
"loss": 3.4544,
"step": 251000
},
{
"epoch": 0.94,
"learning_rate": 3.0699528273720668e-06,
"loss": 3.4487,
"step": 251500
},
{
"epoch": 0.94,
"learning_rate": 2.976652534782349e-06,
"loss": 3.4497,
"step": 252000
},
{
"epoch": 0.94,
"learning_rate": 2.8833522421926314e-06,
"loss": 3.4464,
"step": 252500
},
{
"epoch": 0.94,
"learning_rate": 2.790051949602914e-06,
"loss": 3.4464,
"step": 253000
},
{
"epoch": 0.95,
"learning_rate": 2.6967516570131964e-06,
"loss": 3.4446,
"step": 253500
},
{
"epoch": 0.95,
"learning_rate": 2.603451364423479e-06,
"loss": 3.4536,
"step": 254000
},
{
"epoch": 0.95,
"learning_rate": 2.5101510718337614e-06,
"loss": 3.4477,
"step": 254500
},
{
"epoch": 0.95,
"learning_rate": 2.416850779244044e-06,
"loss": 3.4474,
"step": 255000
},
{
"epoch": 0.95,
"learning_rate": 2.3235504866543264e-06,
"loss": 3.4475,
"step": 255500
},
{
"epoch": 0.96,
"learning_rate": 2.2302501940646087e-06,
"loss": 3.449,
"step": 256000
},
{
"epoch": 0.96,
"learning_rate": 2.136949901474891e-06,
"loss": 3.4525,
"step": 256500
},
{
"epoch": 0.96,
"learning_rate": 2.0436496088851733e-06,
"loss": 3.4448,
"step": 257000
},
{
"epoch": 0.96,
"learning_rate": 1.950349316295456e-06,
"loss": 3.4487,
"step": 257500
},
{
"epoch": 0.96,
"learning_rate": 1.8570490237057383e-06,
"loss": 3.4372,
"step": 258000
},
{
"epoch": 0.96,
"learning_rate": 1.763748731116021e-06,
"loss": 3.4422,
"step": 258500
},
{
"epoch": 0.97,
"learning_rate": 1.6704484385263034e-06,
"loss": 3.4483,
"step": 259000
},
{
"epoch": 0.97,
"learning_rate": 1.5771481459365859e-06,
"loss": 3.4515,
"step": 259500
},
{
"epoch": 0.97,
"learning_rate": 1.4838478533468682e-06,
"loss": 3.4508,
"step": 260000
},
{
"epoch": 0.97,
"eval_accuracy": 0.39486900634888356,
"eval_loss": 3.367981433868408,
"eval_runtime": 6161.9743,
"eval_samples_per_second": 73.221,
"eval_steps_per_second": 1.144,
"step": 260000
},
{
"epoch": 0.97,
"learning_rate": 1.3905475607571507e-06,
"loss": 3.4452,
"step": 260500
},
{
"epoch": 0.97,
"learning_rate": 1.297247268167433e-06,
"loss": 3.445,
"step": 261000
},
{
"epoch": 0.98,
"learning_rate": 1.2039469755777155e-06,
"loss": 3.454,
"step": 261500
},
{
"epoch": 0.98,
"learning_rate": 1.1106466829879978e-06,
"loss": 3.4487,
"step": 262000
},
{
"epoch": 0.98,
"learning_rate": 1.0173463903982803e-06,
"loss": 3.4454,
"step": 262500
},
{
"epoch": 0.98,
"learning_rate": 9.240460978085627e-07,
"loss": 3.4432,
"step": 263000
},
{
"epoch": 0.98,
"learning_rate": 8.307458052188452e-07,
"loss": 3.447,
"step": 263500
},
{
"epoch": 0.99,
"learning_rate": 7.374455126291276e-07,
"loss": 3.4449,
"step": 264000
},
{
"epoch": 0.99,
"learning_rate": 6.4414522003941e-07,
"loss": 3.4486,
"step": 264500
},
{
"epoch": 0.99,
"learning_rate": 5.508449274496924e-07,
"loss": 3.4444,
"step": 265000
},
{
"epoch": 0.99,
"learning_rate": 4.5754463485997496e-07,
"loss": 3.4486,
"step": 265500
},
{
"epoch": 0.99,
"learning_rate": 3.6424434227025736e-07,
"loss": 3.4416,
"step": 266000
},
{
"epoch": 0.99,
"learning_rate": 2.709440496805398e-07,
"loss": 3.4419,
"step": 266500
},
{
"epoch": 1.0,
"learning_rate": 1.7764375709082225e-07,
"loss": 3.4406,
"step": 267000
},
{
"epoch": 1.0,
"learning_rate": 8.434346450110468e-08,
"loss": 3.4406,
"step": 267500
},
{
"epoch": 1.0,
"step": 267952,
"total_flos": 4.480871972732928e+18,
"train_loss": 3.648317173601172,
"train_runtime": 313365.2222,
"train_samples_per_second": 27.362,
"train_steps_per_second": 0.855
}
],
"logging_steps": 500,
"max_steps": 267952,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 10000,
"total_flos": 4.480871972732928e+18,
"train_batch_size": 32,
"trial_name": null,
"trial_params": null
}