{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.99949083503055,
  "eval_steps": 100,
  "global_step": 5891,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.025458248472505093,
      "grad_norm": 0.7198461890220642,
      "learning_rate": 1e-05,
      "loss": 0.2893,
      "step": 50
    },
    {
      "epoch": 0.05091649694501019,
      "grad_norm": 1.0035477876663208,
      "learning_rate": 1e-05,
      "loss": 0.2938,
      "step": 100
    },
    {
      "epoch": 0.05091649694501019,
      "eval_loss": 0.30817005038261414,
      "eval_runtime": 22.8318,
      "eval_samples_per_second": 4.38,
      "eval_steps_per_second": 0.569,
      "step": 100
    },
    {
      "epoch": 0.07637474541751528,
      "grad_norm": 1.1024292707443237,
      "learning_rate": 1e-05,
      "loss": 0.3017,
      "step": 150
    },
    {
      "epoch": 0.10183299389002037,
      "grad_norm": 0.8950141668319702,
      "learning_rate": 1e-05,
      "loss": 0.2912,
      "step": 200
    },
    {
      "epoch": 0.10183299389002037,
      "eval_loss": 0.30671748518943787,
      "eval_runtime": 21.6577,
      "eval_samples_per_second": 4.617,
      "eval_steps_per_second": 0.6,
      "step": 200
    },
    {
      "epoch": 0.12729124236252545,
      "grad_norm": 0.9789568781852722,
      "learning_rate": 1e-05,
      "loss": 0.2928,
      "step": 250
    },
    {
      "epoch": 0.15274949083503056,
      "grad_norm": 1.0154632329940796,
      "learning_rate": 1e-05,
      "loss": 0.2839,
      "step": 300
    },
    {
      "epoch": 0.15274949083503056,
      "eval_loss": 0.30376389622688293,
      "eval_runtime": 21.5085,
      "eval_samples_per_second": 4.649,
      "eval_steps_per_second": 0.604,
      "step": 300
    },
    {
      "epoch": 0.17820773930753564,
      "grad_norm": 0.977684736251831,
      "learning_rate": 1e-05,
      "loss": 0.299,
      "step": 350
    },
    {
      "epoch": 0.20366598778004075,
      "grad_norm": 1.02386474609375,
      "learning_rate": 1e-05,
      "loss": 0.2922,
      "step": 400
    },
    {
      "epoch": 0.20366598778004075,
      "eval_loss": 0.3035086393356323,
      "eval_runtime": 21.9907,
      "eval_samples_per_second": 4.547,
      "eval_steps_per_second": 0.591,
      "step": 400
    },
    {
      "epoch": 0.22912423625254583,
      "grad_norm": 1.0174798965454102,
      "learning_rate": 1e-05,
      "loss": 0.2955,
      "step": 450
    },
    {
      "epoch": 0.2545824847250509,
      "grad_norm": 1.0312519073486328,
      "learning_rate": 1e-05,
      "loss": 0.3013,
      "step": 500
    },
    {
      "epoch": 0.2545824847250509,
      "eval_loss": 0.29991400241851807,
      "eval_runtime": 21.9988,
      "eval_samples_per_second": 4.546,
      "eval_steps_per_second": 0.591,
      "step": 500
    },
    {
      "epoch": 0.280040733197556,
      "grad_norm": 0.8153128623962402,
      "learning_rate": 1e-05,
      "loss": 0.2902,
      "step": 550
    },
    {
      "epoch": 0.3054989816700611,
      "grad_norm": 0.9280871748924255,
      "learning_rate": 1e-05,
      "loss": 0.2933,
      "step": 600
    },
    {
      "epoch": 0.3054989816700611,
      "eval_loss": 0.29873424768447876,
      "eval_runtime": 21.424,
      "eval_samples_per_second": 4.668,
      "eval_steps_per_second": 0.607,
      "step": 600
    },
    {
      "epoch": 0.33095723014256617,
      "grad_norm": 1.0311402082443237,
      "learning_rate": 1e-05,
      "loss": 0.2871,
      "step": 650
    },
    {
      "epoch": 0.3564154786150713,
      "grad_norm": 1.1811566352844238,
      "learning_rate": 1e-05,
      "loss": 0.2968,
      "step": 700
    },
    {
      "epoch": 0.3564154786150713,
      "eval_loss": 0.29955142736434937,
      "eval_runtime": 23.1012,
      "eval_samples_per_second": 4.329,
      "eval_steps_per_second": 0.563,
      "step": 700
    },
    {
      "epoch": 0.3818737270875764,
      "grad_norm": 1.2003265619277954,
      "learning_rate": 1e-05,
      "loss": 0.2887,
      "step": 750
    },
    {
      "epoch": 0.4073319755600815,
      "grad_norm": 1.488318920135498,
      "learning_rate": 1e-05,
      "loss": 0.2891,
      "step": 800
    },
    {
      "epoch": 0.4073319755600815,
      "eval_loss": 0.29836124181747437,
      "eval_runtime": 22.3121,
      "eval_samples_per_second": 4.482,
      "eval_steps_per_second": 0.583,
      "step": 800
    },
    {
      "epoch": 0.43279022403258655,
      "grad_norm": 0.858514130115509,
      "learning_rate": 1e-05,
      "loss": 0.2985,
      "step": 850
    },
    {
      "epoch": 0.45824847250509165,
      "grad_norm": 0.9991205930709839,
      "learning_rate": 1e-05,
      "loss": 0.2877,
      "step": 900
    },
    {
      "epoch": 0.45824847250509165,
      "eval_loss": 0.2956816554069519,
      "eval_runtime": 21.6536,
      "eval_samples_per_second": 4.618,
      "eval_steps_per_second": 0.6,
      "step": 900
    },
    {
      "epoch": 0.48370672097759676,
      "grad_norm": 1.130011796951294,
      "learning_rate": 1e-05,
      "loss": 0.2831,
      "step": 950
    },
    {
      "epoch": 0.5091649694501018,
      "grad_norm": 0.97832852602005,
      "learning_rate": 1e-05,
      "loss": 0.2844,
      "step": 1000
    },
    {
      "epoch": 0.5091649694501018,
      "eval_loss": 0.29429855942726135,
      "eval_runtime": 21.7902,
      "eval_samples_per_second": 4.589,
      "eval_steps_per_second": 0.597,
      "step": 1000
    },
    {
      "epoch": 0.5346232179226069,
      "grad_norm": 1.1608392000198364,
      "learning_rate": 1e-05,
      "loss": 0.2802,
      "step": 1050
    },
    {
      "epoch": 0.560081466395112,
      "grad_norm": 0.9106999635696411,
      "learning_rate": 1e-05,
      "loss": 0.2736,
      "step": 1100
    },
    {
      "epoch": 0.560081466395112,
      "eval_loss": 0.29138484597206116,
      "eval_runtime": 21.847,
      "eval_samples_per_second": 4.577,
      "eval_steps_per_second": 0.595,
      "step": 1100
    },
    {
      "epoch": 0.5855397148676171,
      "grad_norm": 1.077606201171875,
      "learning_rate": 1e-05,
      "loss": 0.2916,
      "step": 1150
    },
    {
      "epoch": 0.6109979633401222,
      "grad_norm": 1.078594446182251,
      "learning_rate": 1e-05,
      "loss": 0.2877,
      "step": 1200
    },
    {
      "epoch": 0.6109979633401222,
      "eval_loss": 0.2930351495742798,
      "eval_runtime": 21.8584,
      "eval_samples_per_second": 4.575,
      "eval_steps_per_second": 0.595,
      "step": 1200
    },
    {
      "epoch": 0.6364562118126272,
      "grad_norm": 1.044995665550232,
      "learning_rate": 1e-05,
      "loss": 0.2852,
      "step": 1250
    },
    {
      "epoch": 0.6619144602851323,
      "grad_norm": 1.0691392421722412,
      "learning_rate": 1e-05,
      "loss": 0.2914,
      "step": 1300
    },
    {
      "epoch": 0.6619144602851323,
      "eval_loss": 0.29031333327293396,
      "eval_runtime": 21.8883,
      "eval_samples_per_second": 4.569,
      "eval_steps_per_second": 0.594,
      "step": 1300
    },
    {
      "epoch": 0.6873727087576375,
      "grad_norm": 1.165562629699707,
      "learning_rate": 1e-05,
      "loss": 0.2854,
      "step": 1350
    },
    {
      "epoch": 0.7128309572301426,
      "grad_norm": 1.1224968433380127,
      "learning_rate": 1e-05,
      "loss": 0.2846,
      "step": 1400
    },
    {
      "epoch": 0.7128309572301426,
      "eval_loss": 0.2897338271141052,
      "eval_runtime": 22.173,
      "eval_samples_per_second": 4.51,
      "eval_steps_per_second": 0.586,
      "step": 1400
    },
    {
      "epoch": 0.7382892057026477,
      "grad_norm": 0.9951677918434143,
      "learning_rate": 1e-05,
      "loss": 0.2754,
      "step": 1450
    },
    {
      "epoch": 0.7637474541751528,
      "grad_norm": 1.116921305656433,
      "learning_rate": 1e-05,
      "loss": 0.2754,
      "step": 1500
    },
    {
      "epoch": 0.7637474541751528,
      "eval_loss": 0.2867298424243927,
      "eval_runtime": 21.7838,
      "eval_samples_per_second": 4.591,
      "eval_steps_per_second": 0.597,
      "step": 1500
    },
    {
      "epoch": 0.7892057026476579,
      "grad_norm": 1.104265570640564,
      "learning_rate": 1e-05,
      "loss": 0.2824,
      "step": 1550
    },
    {
      "epoch": 0.814663951120163,
      "grad_norm": 0.8793336749076843,
      "learning_rate": 1e-05,
      "loss": 0.2901,
      "step": 1600
    },
    {
      "epoch": 0.814663951120163,
      "eval_loss": 0.28652000427246094,
      "eval_runtime": 21.8736,
      "eval_samples_per_second": 4.572,
      "eval_steps_per_second": 0.594,
      "step": 1600
    },
    {
      "epoch": 0.840122199592668,
      "grad_norm": 1.2304877042770386,
      "learning_rate": 1e-05,
      "loss": 0.2718,
      "step": 1650
    },
    {
      "epoch": 0.8655804480651731,
      "grad_norm": 0.9079441428184509,
      "learning_rate": 1e-05,
      "loss": 0.2787,
      "step": 1700
    },
    {
      "epoch": 0.8655804480651731,
      "eval_loss": 0.2839984893798828,
      "eval_runtime": 21.7844,
      "eval_samples_per_second": 4.59,
      "eval_steps_per_second": 0.597,
      "step": 1700
    },
    {
      "epoch": 0.8910386965376782,
      "grad_norm": 1.359052300453186,
      "learning_rate": 1e-05,
      "loss": 0.2703,
      "step": 1750
    },
    {
      "epoch": 0.9164969450101833,
      "grad_norm": 1.0245873928070068,
      "learning_rate": 1e-05,
      "loss": 0.2678,
      "step": 1800
    },
    {
      "epoch": 0.9164969450101833,
      "eval_loss": 0.2825533151626587,
      "eval_runtime": 22.022,
      "eval_samples_per_second": 4.541,
      "eval_steps_per_second": 0.59,
      "step": 1800
    },
    {
      "epoch": 0.9419551934826884,
      "grad_norm": 0.9011121988296509,
      "learning_rate": 1e-05,
      "loss": 0.2747,
      "step": 1850
    },
    {
      "epoch": 0.9674134419551935,
      "grad_norm": 1.006032943725586,
      "learning_rate": 1e-05,
      "loss": 0.2721,
      "step": 1900
    },
    {
      "epoch": 0.9674134419551935,
      "eval_loss": 0.2824758291244507,
      "eval_runtime": 22.1289,
      "eval_samples_per_second": 4.519,
      "eval_steps_per_second": 0.587,
      "step": 1900
    },
    {
      "epoch": 0.9928716904276986,
      "grad_norm": 0.9993594288825989,
      "learning_rate": 1e-05,
      "loss": 0.2784,
      "step": 1950
    },
    {
      "epoch": 1.0183299389002036,
      "grad_norm": 0.9710284471511841,
      "learning_rate": 1e-05,
      "loss": 0.2713,
      "step": 2000
    },
    {
      "epoch": 1.0183299389002036,
      "eval_loss": 0.2820639908313751,
      "eval_runtime": 21.7507,
      "eval_samples_per_second": 4.598,
      "eval_steps_per_second": 0.598,
      "step": 2000
    },
    {
      "epoch": 1.0437881873727088,
      "grad_norm": 1.2046750783920288,
      "learning_rate": 1e-05,
      "loss": 0.2799,
      "step": 2050
    },
    {
      "epoch": 1.0692464358452138,
      "grad_norm": 0.9969730377197266,
      "learning_rate": 1e-05,
      "loss": 0.2842,
      "step": 2100
    },
    {
      "epoch": 1.0692464358452138,
      "eval_loss": 0.28064805269241333,
      "eval_runtime": 21.7423,
      "eval_samples_per_second": 4.599,
      "eval_steps_per_second": 0.598,
      "step": 2100
    },
    {
      "epoch": 1.094704684317719,
      "grad_norm": 0.9368526339530945,
      "learning_rate": 1e-05,
      "loss": 0.2783,
      "step": 2150
    },
    {
      "epoch": 1.120162932790224,
      "grad_norm": 1.2995036840438843,
      "learning_rate": 1e-05,
      "loss": 0.2781,
      "step": 2200
    },
    {
      "epoch": 1.120162932790224,
      "eval_loss": 0.2789928913116455,
      "eval_runtime": 21.6436,
      "eval_samples_per_second": 4.62,
      "eval_steps_per_second": 0.601,
      "step": 2200
    },
    {
      "epoch": 1.145621181262729,
      "grad_norm": 1.2737852334976196,
      "learning_rate": 1e-05,
      "loss": 0.2731,
      "step": 2250
    },
    {
      "epoch": 1.1710794297352343,
      "grad_norm": 1.0202410221099854,
      "learning_rate": 1e-05,
      "loss": 0.273,
      "step": 2300
    },
    {
      "epoch": 1.1710794297352343,
      "eval_loss": 0.27778831124305725,
      "eval_runtime": 21.9156,
      "eval_samples_per_second": 4.563,
      "eval_steps_per_second": 0.593,
      "step": 2300
    },
    {
      "epoch": 1.1965376782077393,
      "grad_norm": 0.9710997939109802,
      "learning_rate": 1e-05,
      "loss": 0.2578,
      "step": 2350
    },
    {
      "epoch": 1.2219959266802445,
      "grad_norm": 0.86209636926651,
      "learning_rate": 1e-05,
      "loss": 0.2714,
      "step": 2400
    },
    {
      "epoch": 1.2219959266802445,
      "eval_loss": 0.27559801936149597,
      "eval_runtime": 21.9259,
      "eval_samples_per_second": 4.561,
      "eval_steps_per_second": 0.593,
      "step": 2400
    },
    {
      "epoch": 1.2474541751527495,
      "grad_norm": 1.0652376413345337,
      "learning_rate": 1e-05,
      "loss": 0.265,
      "step": 2450
    },
    {
      "epoch": 1.2729124236252547,
      "grad_norm": 1.002944827079773,
      "learning_rate": 1e-05,
      "loss": 0.2536,
      "step": 2500
    },
    {
      "epoch": 1.2729124236252547,
      "eval_loss": 0.2747356593608856,
      "eval_runtime": 21.9091,
      "eval_samples_per_second": 4.564,
      "eval_steps_per_second": 0.593,
      "step": 2500
    },
    {
      "epoch": 1.2983706720977597,
      "grad_norm": 1.0743255615234375,
      "learning_rate": 1e-05,
      "loss": 0.2742,
      "step": 2550
    },
    {
      "epoch": 1.3238289205702647,
      "grad_norm": 1.2174959182739258,
      "learning_rate": 1e-05,
      "loss": 0.268,
      "step": 2600
    },
    {
      "epoch": 1.3238289205702647,
      "eval_loss": 0.27502280473709106,
      "eval_runtime": 21.9371,
      "eval_samples_per_second": 4.558,
      "eval_steps_per_second": 0.593,
      "step": 2600
    },
    {
      "epoch": 1.34928716904277,
      "grad_norm": 1.1314553022384644,
      "learning_rate": 1e-05,
      "loss": 0.26,
      "step": 2650
    },
    {
      "epoch": 1.374745417515275,
      "grad_norm": 1.007804036140442,
      "learning_rate": 1e-05,
      "loss": 0.2776,
      "step": 2700
    },
    {
      "epoch": 1.374745417515275,
      "eval_loss": 0.27191075682640076,
      "eval_runtime": 21.9972,
      "eval_samples_per_second": 4.546,
      "eval_steps_per_second": 0.591,
      "step": 2700
    },
    {
      "epoch": 1.4002036659877801,
      "grad_norm": 1.1400426626205444,
      "learning_rate": 1e-05,
      "loss": 0.2577,
      "step": 2750
    },
    {
      "epoch": 1.4256619144602851,
      "grad_norm": 0.9301505088806152,
      "learning_rate": 1e-05,
      "loss": 0.2726,
      "step": 2800
    },
    {
      "epoch": 1.4256619144602851,
      "eval_loss": 0.2725023925304413,
      "eval_runtime": 21.722,
      "eval_samples_per_second": 4.604,
      "eval_steps_per_second": 0.598,
      "step": 2800
    },
    {
      "epoch": 1.4511201629327903,
      "grad_norm": 1.142259120941162,
      "learning_rate": 1e-05,
      "loss": 0.2635,
      "step": 2850
    },
    {
      "epoch": 1.4765784114052953,
      "grad_norm": 1.1009142398834229,
      "learning_rate": 1e-05,
      "loss": 0.2572,
      "step": 2900
    },
    {
      "epoch": 1.4765784114052953,
      "eval_loss": 0.2723616063594818,
      "eval_runtime": 21.8662,
      "eval_samples_per_second": 4.573,
      "eval_steps_per_second": 0.595,
      "step": 2900
    },
    {
      "epoch": 1.5020366598778003,
      "grad_norm": 1.2985098361968994,
      "learning_rate": 1e-05,
      "loss": 0.2477,
      "step": 2950
    },
    {
      "epoch": 1.5274949083503055,
      "grad_norm": 1.2853788137435913,
      "learning_rate": 1e-05,
      "loss": 0.2518,
      "step": 3000
    },
    {
      "epoch": 1.5274949083503055,
      "eval_loss": 0.2711648643016815,
      "eval_runtime": 22.1089,
      "eval_samples_per_second": 4.523,
      "eval_steps_per_second": 0.588,
      "step": 3000
    },
    {
      "epoch": 1.5529531568228105,
      "grad_norm": 1.3646196126937866,
      "learning_rate": 1e-05,
      "loss": 0.2545,
      "step": 3050
    },
    {
      "epoch": 1.5784114052953155,
      "grad_norm": 1.0580254793167114,
      "learning_rate": 1e-05,
      "loss": 0.2665,
      "step": 3100
    },
    {
      "epoch": 1.5784114052953155,
      "eval_loss": 0.27152860164642334,
      "eval_runtime": 21.563,
      "eval_samples_per_second": 4.638,
      "eval_steps_per_second": 0.603,
      "step": 3100
    },
    {
      "epoch": 1.6038696537678208,
      "grad_norm": 1.529466152191162,
      "learning_rate": 1e-05,
      "loss": 0.2589,
      "step": 3150
    },
    {
      "epoch": 1.629327902240326,
      "grad_norm": 1.0657099485397339,
      "learning_rate": 1e-05,
      "loss": 0.2559,
      "step": 3200
    },
    {
      "epoch": 1.629327902240326,
      "eval_loss": 0.2696399390697479,
      "eval_runtime": 21.6487,
      "eval_samples_per_second": 4.619,
      "eval_steps_per_second": 0.6,
      "step": 3200
    },
    {
      "epoch": 1.654786150712831,
      "grad_norm": 1.0233310461044312,
      "learning_rate": 1e-05,
      "loss": 0.2654,
      "step": 3250
    },
    {
      "epoch": 1.680244399185336,
      "grad_norm": 1.151859998703003,
      "learning_rate": 1e-05,
      "loss": 0.2609,
      "step": 3300
    },
    {
      "epoch": 1.680244399185336,
      "eval_loss": 0.26971447467803955,
      "eval_runtime": 21.8279,
      "eval_samples_per_second": 4.581,
      "eval_steps_per_second": 0.596,
      "step": 3300
    },
    {
      "epoch": 1.7057026476578412,
      "grad_norm": 1.1169921159744263,
      "learning_rate": 1e-05,
      "loss": 0.266,
      "step": 3350
    },
    {
      "epoch": 1.7311608961303462,
      "grad_norm": 0.9375786185264587,
      "learning_rate": 1e-05,
      "loss": 0.2603,
      "step": 3400
    },
    {
      "epoch": 1.7311608961303462,
      "eval_loss": 0.26760581135749817,
      "eval_runtime": 21.6655,
      "eval_samples_per_second": 4.616,
      "eval_steps_per_second": 0.6,
      "step": 3400
    },
    {
      "epoch": 1.7566191446028512,
      "grad_norm": 1.0559574365615845,
      "learning_rate": 1e-05,
      "loss": 0.2644,
      "step": 3450
    },
    {
      "epoch": 1.7820773930753564,
      "grad_norm": 1.1326260566711426,
      "learning_rate": 1e-05,
      "loss": 0.2448,
      "step": 3500
    },
    {
      "epoch": 1.7820773930753564,
      "eval_loss": 0.2665054500102997,
      "eval_runtime": 21.6484,
      "eval_samples_per_second": 4.619,
      "eval_steps_per_second": 0.601,
      "step": 3500
    },
    {
      "epoch": 1.8075356415478616,
      "grad_norm": 1.396811842918396,
      "learning_rate": 1e-05,
      "loss": 0.252,
      "step": 3550
    },
    {
      "epoch": 1.8329938900203666,
      "grad_norm": 1.6235796213150024,
      "learning_rate": 1e-05,
      "loss": 0.2587,
      "step": 3600
    },
    {
      "epoch": 1.8329938900203666,
      "eval_loss": 0.2665034234523773,
      "eval_runtime": 21.9855,
      "eval_samples_per_second": 4.548,
      "eval_steps_per_second": 0.591,
      "step": 3600
    },
    {
      "epoch": 1.8584521384928716,
      "grad_norm": 1.2640048265457153,
      "learning_rate": 1e-05,
      "loss": 0.2525,
      "step": 3650
    },
    {
      "epoch": 1.8839103869653768,
      "grad_norm": 1.1669272184371948,
      "learning_rate": 1e-05,
      "loss": 0.2534,
      "step": 3700
    },
    {
      "epoch": 1.8839103869653768,
      "eval_loss": 0.26496145129203796,
      "eval_runtime": 22.0324,
      "eval_samples_per_second": 4.539,
      "eval_steps_per_second": 0.59,
      "step": 3700
    },
    {
      "epoch": 1.9093686354378818,
      "grad_norm": 1.2015262842178345,
      "learning_rate": 1e-05,
      "loss": 0.2532,
      "step": 3750
    },
    {
      "epoch": 1.9348268839103868,
      "grad_norm": 1.0683043003082275,
      "learning_rate": 1e-05,
      "loss": 0.2496,
      "step": 3800
    },
    {
      "epoch": 1.9348268839103868,
      "eval_loss": 0.2648490369319916,
      "eval_runtime": 21.9752,
      "eval_samples_per_second": 4.551,
      "eval_steps_per_second": 0.592,
      "step": 3800
    },
    {
      "epoch": 1.960285132382892,
      "grad_norm": 1.4562475681304932,
      "learning_rate": 1e-05,
      "loss": 0.242,
      "step": 3850
    },
    {
      "epoch": 1.9857433808553973,
      "grad_norm": 0.9929770231246948,
      "learning_rate": 1e-05,
      "loss": 0.2528,
      "step": 3900
    },
    {
      "epoch": 1.9857433808553973,
      "eval_loss": 0.263571560382843,
      "eval_runtime": 21.7216,
      "eval_samples_per_second": 4.604,
      "eval_steps_per_second": 0.598,
      "step": 3900
    },
    {
      "epoch": 2.011201629327902,
      "grad_norm": 1.0417041778564453,
      "learning_rate": 1e-05,
      "loss": 0.2452,
      "step": 3950
    },
    {
      "epoch": 2.0366598778004072,
      "grad_norm": 1.0510022640228271,
      "learning_rate": 1e-05,
      "loss": 0.2652,
      "step": 4000
    },
    {
      "epoch": 2.0366598778004072,
      "eval_loss": 0.262724906206131,
      "eval_runtime": 22.0633,
      "eval_samples_per_second": 4.532,
      "eval_steps_per_second": 0.589,
      "step": 4000
    },
    {
      "epoch": 2.0621181262729125,
      "grad_norm": 1.383092999458313,
      "learning_rate": 1e-05,
      "loss": 0.2369,
      "step": 4050
    },
    {
      "epoch": 2.0875763747454177,
      "grad_norm": 1.3613831996917725,
      "learning_rate": 1e-05,
      "loss": 0.2493,
      "step": 4100
    },
    {
      "epoch": 2.0875763747454177,
      "eval_loss": 0.2625581622123718,
      "eval_runtime": 21.7251,
      "eval_samples_per_second": 4.603,
      "eval_steps_per_second": 0.598,
      "step": 4100
    },
    {
      "epoch": 2.1130346232179225,
      "grad_norm": 1.0267040729522705,
      "learning_rate": 1e-05,
      "loss": 0.2511,
      "step": 4150
    },
    {
      "epoch": 2.1384928716904277,
      "grad_norm": 0.9748584032058716,
      "learning_rate": 1e-05,
      "loss": 0.2383,
      "step": 4200
    },
    {
      "epoch": 2.1384928716904277,
      "eval_loss": 0.26004984974861145,
      "eval_runtime": 21.6797,
      "eval_samples_per_second": 4.613,
      "eval_steps_per_second": 0.6,
      "step": 4200
    },
    {
      "epoch": 2.163951120162933,
      "grad_norm": 1.3859432935714722,
      "learning_rate": 1e-05,
      "loss": 0.2544,
      "step": 4250
    },
    {
      "epoch": 2.189409368635438,
      "grad_norm": 1.598718285560608,
      "learning_rate": 1e-05,
      "loss": 0.2527,
      "step": 4300
    },
    {
      "epoch": 2.189409368635438,
      "eval_loss": 0.25936272740364075,
      "eval_runtime": 21.803,
      "eval_samples_per_second": 4.587,
      "eval_steps_per_second": 0.596,
      "step": 4300
    },
    {
      "epoch": 2.214867617107943,
      "grad_norm": 0.9896050691604614,
      "learning_rate": 1e-05,
      "loss": 0.2423,
      "step": 4350
    },
    {
      "epoch": 2.240325865580448,
      "grad_norm": 1.4134578704833984,
      "learning_rate": 1e-05,
      "loss": 0.2446,
      "step": 4400
    },
    {
      "epoch": 2.240325865580448,
      "eval_loss": 0.2597595751285553,
      "eval_runtime": 21.9078,
      "eval_samples_per_second": 4.565,
      "eval_steps_per_second": 0.593,
      "step": 4400
    },
    {
      "epoch": 2.2657841140529533,
      "grad_norm": 1.1840572357177734,
      "learning_rate": 1e-05,
      "loss": 0.251,
      "step": 4450
    },
    {
      "epoch": 2.291242362525458,
      "grad_norm": 1.3326150178909302,
      "learning_rate": 1e-05,
      "loss": 0.2504,
      "step": 4500
    },
    {
      "epoch": 2.291242362525458,
      "eval_loss": 0.2583908438682556,
      "eval_runtime": 21.7836,
      "eval_samples_per_second": 4.591,
      "eval_steps_per_second": 0.597,
      "step": 4500
    },
    {
      "epoch": 2.3167006109979633,
      "grad_norm": 1.4150619506835938,
      "learning_rate": 1e-05,
      "loss": 0.2481,
      "step": 4550
    },
    {
      "epoch": 2.3421588594704685,
      "grad_norm": 1.4056681394577026,
      "learning_rate": 1e-05,
      "loss": 0.2474,
      "step": 4600
    },
    {
      "epoch": 2.3421588594704685,
      "eval_loss": 0.25896069407463074,
      "eval_runtime": 21.9165,
      "eval_samples_per_second": 4.563,
      "eval_steps_per_second": 0.593,
      "step": 4600
    },
    {
      "epoch": 2.3676171079429738,
      "grad_norm": 1.1109027862548828,
      "learning_rate": 1e-05,
      "loss": 0.2447,
      "step": 4650
    },
    {
      "epoch": 2.3930753564154785,
      "grad_norm": 1.1159225702285767,
      "learning_rate": 1e-05,
      "loss": 0.2482,
      "step": 4700
    },
    {
      "epoch": 2.3930753564154785,
      "eval_loss": 0.2578243911266327,
      "eval_runtime": 21.5943,
      "eval_samples_per_second": 4.631,
      "eval_steps_per_second": 0.602,
      "step": 4700
    },
    {
      "epoch": 2.4185336048879837,
      "grad_norm": 1.1924070119857788,
      "learning_rate": 1e-05,
      "loss": 0.2506,
      "step": 4750
    },
    {
      "epoch": 2.443991853360489,
      "grad_norm": 1.5512938499450684,
      "learning_rate": 1e-05,
      "loss": 0.2405,
      "step": 4800
    },
    {
      "epoch": 2.443991853360489,
      "eval_loss": 0.2575813829898834,
      "eval_runtime": 22.0278,
      "eval_samples_per_second": 4.54,
      "eval_steps_per_second": 0.59,
      "step": 4800
    },
    {
      "epoch": 2.4694501018329937,
      "grad_norm": 1.0490576028823853,
      "learning_rate": 1e-05,
      "loss": 0.2476,
      "step": 4850
    },
    {
      "epoch": 2.494908350305499,
      "grad_norm": 1.0198191404342651,
      "learning_rate": 1e-05,
      "loss": 0.2399,
      "step": 4900
    },
    {
      "epoch": 2.494908350305499,
      "eval_loss": 0.25637081265449524,
      "eval_runtime": 21.9152,
      "eval_samples_per_second": 4.563,
      "eval_steps_per_second": 0.593,
      "step": 4900
    },
    {
      "epoch": 2.520366598778004,
      "grad_norm": 1.122515320777893,
      "learning_rate": 1e-05,
      "loss": 0.2501,
      "step": 4950
    },
    {
      "epoch": 2.5458248472505094,
      "grad_norm": 1.082686424255371,
      "learning_rate": 1e-05,
      "loss": 0.2443,
      "step": 5000
    },
    {
      "epoch": 2.5458248472505094,
      "eval_loss": 0.25523728132247925,
      "eval_runtime": 21.0107,
      "eval_samples_per_second": 4.759,
      "eval_steps_per_second": 0.619,
      "step": 5000
    },
    {
      "epoch": 2.571283095723014,
      "grad_norm": 1.0135226249694824,
      "learning_rate": 1e-05,
      "loss": 0.2414,
      "step": 5050
    },
    {
      "epoch": 2.5967413441955194,
      "grad_norm": 0.9929371476173401,
      "learning_rate": 1e-05,
      "loss": 0.248,
      "step": 5100
    },
    {
      "epoch": 2.5967413441955194,
      "eval_loss": 0.2532651722431183,
      "eval_runtime": 21.3518,
      "eval_samples_per_second": 4.683,
      "eval_steps_per_second": 0.609,
      "step": 5100
    },
    {
      "epoch": 2.6221995926680246,
      "grad_norm": 1.1128815412521362,
      "learning_rate": 1e-05,
      "loss": 0.2375,
      "step": 5150
    },
    {
      "epoch": 2.6476578411405294,
      "grad_norm": 1.5581951141357422,
      "learning_rate": 1e-05,
      "loss": 0.2432,
      "step": 5200
    },
    {
      "epoch": 2.6476578411405294,
      "eval_loss": 0.2521425485610962,
      "eval_runtime": 21.0088,
      "eval_samples_per_second": 4.76,
      "eval_steps_per_second": 0.619,
      "step": 5200
    },
    {
      "epoch": 2.6731160896130346,
      "grad_norm": 1.1291751861572266,
      "learning_rate": 1e-05,
      "loss": 0.2415,
      "step": 5250
    },
    {
      "epoch": 2.69857433808554,
      "grad_norm": 1.139137625694275,
      "learning_rate": 1e-05,
      "loss": 0.2398,
      "step": 5300
    },
    {
      "epoch": 2.69857433808554,
      "eval_loss": 0.25268828868865967,
      "eval_runtime": 21.7715,
      "eval_samples_per_second": 4.593,
      "eval_steps_per_second": 0.597,
      "step": 5300
    },
    {
      "epoch": 2.724032586558045,
      "grad_norm": 1.0960917472839355,
      "learning_rate": 1e-05,
      "loss": 0.2397,
      "step": 5350
    },
    {
      "epoch": 2.74949083503055,
      "grad_norm": 1.4622306823730469,
      "learning_rate": 1e-05,
      "loss": 0.2371,
      "step": 5400
    },
    {
      "epoch": 2.74949083503055,
      "eval_loss": 0.25145551562309265,
      "eval_runtime": 21.3534,
      "eval_samples_per_second": 4.683,
      "eval_steps_per_second": 0.609,
      "step": 5400
    },
    {
      "epoch": 2.774949083503055,
      "grad_norm": 1.0508596897125244,
      "learning_rate": 1e-05,
      "loss": 0.2389,
      "step": 5450
    },
    {
      "epoch": 2.8004073319755602,
      "grad_norm": 0.8241857886314392,
      "learning_rate": 1e-05,
      "loss": 0.2403,
      "step": 5500
    },
    {
      "epoch": 2.8004073319755602,
      "eval_loss": 0.25168588757514954,
      "eval_runtime": 21.2395,
      "eval_samples_per_second": 4.708,
      "eval_steps_per_second": 0.612,
      "step": 5500
    },
    {
      "epoch": 2.825865580448065,
      "grad_norm": 1.214141845703125,
      "learning_rate": 1e-05,
      "loss": 0.2406,
      "step": 5550
    },
    {
      "epoch": 2.8513238289205702,
      "grad_norm": 1.3861531019210815,
      "learning_rate": 1e-05,
      "loss": 0.2479,
      "step": 5600
    },
    {
      "epoch": 2.8513238289205702,
      "eval_loss": 0.24935197830200195,
      "eval_runtime": 21.4872,
      "eval_samples_per_second": 4.654,
      "eval_steps_per_second": 0.605,
      "step": 5600
    },
    {
      "epoch": 2.8767820773930755,
      "grad_norm": 1.1861828565597534,
      "learning_rate": 1e-05,
      "loss": 0.2318,
      "step": 5650
    },
    {
      "epoch": 2.9022403258655807,
      "grad_norm": 1.1281813383102417,
      "learning_rate": 1e-05,
      "loss": 0.2413,
      "step": 5700
    },
    {
      "epoch": 2.9022403258655807,
      "eval_loss": 0.25039103627204895,
      "eval_runtime": 20.9675,
      "eval_samples_per_second": 4.769,
      "eval_steps_per_second": 0.62,
      "step": 5700
    },
    {
      "epoch": 2.9276985743380854,
      "grad_norm": 1.1463491916656494,
      "learning_rate": 1e-05,
      "loss": 0.2364,
      "step": 5750
    },
    {
      "epoch": 2.9531568228105907,
      "grad_norm": 1.3118984699249268,
      "learning_rate": 1e-05,
      "loss": 0.2388,
      "step": 5800
    },
    {
      "epoch": 2.9531568228105907,
      "eval_loss": 0.24982962012290955,
      "eval_runtime": 21.5908,
      "eval_samples_per_second": 4.632,
      "eval_steps_per_second": 0.602,
      "step": 5800
    },
    {
      "epoch": 2.978615071283096,
      "grad_norm": 1.0851026773452759,
      "learning_rate": 1e-05,
      "loss": 0.2348,
      "step": 5850
    }
  ],
  "logging_steps": 50,
  "max_steps": 5891,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 6.421089311391744e+20,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}