{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.4696485623003195,
  "eval_steps": 39,
  "global_step": 117,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "grad_norm": 0.06435385346412659,
      "learning_rate": 1e-05,
      "loss": 0.6574,
      "step": 1
    },
    {
      "epoch": 0.01,
      "eval_loss": 0.7461566925048828,
      "eval_runtime": 151.7973,
      "eval_samples_per_second": 1.482,
      "eval_steps_per_second": 0.376,
      "step": 1
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.062134526669979095,
      "learning_rate": 2e-05,
      "loss": 0.6897,
      "step": 2
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.06962593644857407,
      "learning_rate": 3e-05,
      "loss": 0.6974,
      "step": 3
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.0730314701795578,
      "learning_rate": 4e-05,
      "loss": 0.7454,
      "step": 4
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.07659862190485,
      "learning_rate": 5e-05,
      "loss": 0.6962,
      "step": 5
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.07314729690551758,
      "learning_rate": 6e-05,
      "loss": 0.6602,
      "step": 6
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.0892895832657814,
      "learning_rate": 7e-05,
      "loss": 0.6594,
      "step": 7
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.11520648747682571,
      "learning_rate": 8e-05,
      "loss": 0.7406,
      "step": 8
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.1333557516336441,
      "learning_rate": 9e-05,
      "loss": 0.663,
      "step": 9
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.14017663896083832,
      "learning_rate": 0.0001,
      "loss": 0.6185,
      "step": 10
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.1453818678855896,
      "learning_rate": 0.00011000000000000002,
      "loss": 0.602,
      "step": 11
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.14279265701770782,
      "learning_rate": 0.00012,
      "loss": 0.6151,
      "step": 12
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.11934962123632431,
      "learning_rate": 0.00013000000000000002,
      "loss": 0.5085,
      "step": 13
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.11375121772289276,
      "learning_rate": 0.00014,
      "loss": 0.4823,
      "step": 14
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.09707039594650269,
      "learning_rate": 0.00015000000000000001,
      "loss": 0.4281,
      "step": 15
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.11333765089511871,
      "learning_rate": 0.00016,
      "loss": 0.4417,
      "step": 16
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.09926053136587143,
      "learning_rate": 0.00017,
      "loss": 0.3654,
      "step": 17
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.1027929037809372,
      "learning_rate": 0.00018,
      "loss": 0.3748,
      "step": 18
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.09768980741500854,
      "learning_rate": 0.00019,
      "loss": 0.3306,
      "step": 19
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.08072198927402496,
      "learning_rate": 0.0002,
      "loss": 0.355,
      "step": 20
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.06444709748029709,
      "learning_rate": 0.00019997332081116373,
      "loss": 0.3152,
      "step": 21
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.0576409213244915,
      "learning_rate": 0.00019989329748023725,
      "loss": 0.2674,
      "step": 22
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.08624427020549774,
      "learning_rate": 0.0001997599727063717,
      "loss": 0.2807,
      "step": 23
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.08828990906476974,
      "learning_rate": 0.00019957341762950344,
      "loss": 0.2736,
      "step": 24
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.0641990676522255,
      "learning_rate": 0.00019933373179239502,
      "loss": 0.2427,
      "step": 25
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.05487390235066414,
      "learning_rate": 0.0001990410430875205,
      "loss": 0.2737,
      "step": 26
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.0666637197136879,
      "learning_rate": 0.00019869550768882455,
      "loss": 0.2806,
      "step": 27
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.053027719259262085,
      "learning_rate": 0.0001982973099683902,
      "loss": 0.2664,
      "step": 28
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.055217448621988297,
      "learning_rate": 0.0001978466623980609,
      "loss": 0.309,
      "step": 29
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.05493360385298729,
      "learning_rate": 0.0001973438054360693,
      "loss": 0.2318,
      "step": 30
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.0376153439283371,
      "learning_rate": 0.00019678900739873226,
      "loss": 0.2049,
      "step": 31
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.03811287507414818,
      "learning_rate": 0.00019618256431728194,
      "loss": 0.2771,
      "step": 32
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.04185184836387634,
      "learning_rate": 0.000195524799779908,
      "loss": 0.1843,
      "step": 33
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.04743755981326103,
      "learning_rate": 0.0001948160647590966,
      "loss": 0.2414,
      "step": 34
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.03343382850289345,
      "learning_rate": 0.00019405673742435678,
      "loss": 0.2007,
      "step": 35
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.03178093209862709,
      "learning_rate": 0.00019324722294043558,
      "loss": 0.1898,
      "step": 36
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.03482227399945259,
      "learning_rate": 0.0001923879532511287,
      "loss": 0.2657,
      "step": 37
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.03727172687649727,
      "learning_rate": 0.0001914793868488021,
      "loss": 0.2185,
      "step": 38
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.0382206104695797,
      "learning_rate": 0.00019052200852974819,
      "loss": 0.1778,
      "step": 39
    },
    {
      "epoch": 0.5,
      "eval_loss": 0.2122737020254135,
      "eval_runtime": 154.5774,
      "eval_samples_per_second": 1.456,
      "eval_steps_per_second": 0.369,
      "step": 39
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.03256627917289734,
      "learning_rate": 0.00018951632913550626,
      "loss": 0.2528,
      "step": 40
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.032382771372795105,
      "learning_rate": 0.00018846288528028555,
      "loss": 0.2165,
      "step": 41
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.030948858708143234,
      "learning_rate": 0.00018736223906463696,
      "loss": 0.1754,
      "step": 42
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.041598908603191376,
      "learning_rate": 0.00018621497777552507,
      "loss": 0.2676,
      "step": 43
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.03000044636428356,
      "learning_rate": 0.00018502171357296144,
      "loss": 0.1501,
      "step": 44
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.03305242955684662,
      "learning_rate": 0.00018378308316336584,
      "loss": 0.1937,
      "step": 45
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.02798221819102764,
      "learning_rate": 0.00018249974745983023,
      "loss": 0.2241,
      "step": 46
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.03340066224336624,
      "learning_rate": 0.00018117239122946615,
      "loss": 0.2078,
      "step": 47
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.032081808894872665,
      "learning_rate": 0.000179801722728024,
      "loss": 0.2061,
      "step": 48
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.033740073442459106,
      "learning_rate": 0.00017838847332197938,
      "loss": 0.2411,
      "step": 49
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.02972455322742462,
      "learning_rate": 0.00017693339709828792,
      "loss": 0.2346,
      "step": 50
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.03134370595216751,
      "learning_rate": 0.0001754372704620164,
      "loss": 0.2573,
      "step": 51
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.035204820334911346,
      "learning_rate": 0.00017390089172206592,
      "loss": 0.2596,
      "step": 52
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.027202850207686424,
      "learning_rate": 0.00017232508066520702,
      "loss": 0.1738,
      "step": 53
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.02930806204676628,
      "learning_rate": 0.00017071067811865476,
      "loss": 0.2178,
      "step": 54
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.03558573126792908,
      "learning_rate": 0.00016905854550141716,
      "loss": 0.2677,
      "step": 55
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.03179259970784187,
      "learning_rate": 0.00016736956436465573,
      "loss": 0.1704,
      "step": 56
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.0318523608148098,
      "learning_rate": 0.00016564463592130428,
      "loss": 0.2117,
      "step": 57
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.034657321870326996,
      "learning_rate": 0.00016388468056519612,
      "loss": 0.206,
      "step": 58
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.032808613032102585,
      "learning_rate": 0.00016209063737995715,
      "loss": 0.2235,
      "step": 59
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.029593750834465027,
      "learning_rate": 0.00016026346363792567,
      "loss": 0.1976,
      "step": 60
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.033486686646938324,
      "learning_rate": 0.00015840413428936767,
      "loss": 0.184,
      "step": 61
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.029341375455260277,
      "learning_rate": 0.0001565136414422592,
      "loss": 0.2114,
      "step": 62
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.03653452917933464,
      "learning_rate": 0.00015459299383291345,
      "loss": 0.2183,
      "step": 63
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.03146987408399582,
      "learning_rate": 0.0001526432162877356,
      "loss": 0.2444,
      "step": 64
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.03224271163344383,
      "learning_rate": 0.00015066534917639195,
      "loss": 0.2157,
      "step": 65
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.028182610869407654,
      "learning_rate": 0.00014866044785668563,
      "loss": 0.16,
      "step": 66
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.03271663188934326,
      "learning_rate": 0.0001466295821114348,
      "loss": 0.2102,
      "step": 67
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.032980792224407196,
      "learning_rate": 0.00014457383557765386,
      "loss": 0.2605,
      "step": 68
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.03456718847155571,
      "learning_rate": 0.0001424943051683422,
      "loss": 0.212,
      "step": 69
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.036501459777355194,
      "learning_rate": 0.00014039210048718949,
      "loss": 0.1798,
      "step": 70
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.033730726689100266,
      "learning_rate": 0.000138268343236509,
      "loss": 0.2018,
      "step": 71
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.03012872487306595,
      "learning_rate": 0.00013612416661871533,
      "loss": 0.2309,
      "step": 72
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.032879915088415146,
      "learning_rate": 0.00013396071473166613,
      "loss": 0.1912,
      "step": 73
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.02916533872485161,
      "learning_rate": 0.00013177914195819016,
      "loss": 0.2313,
      "step": 74
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.032934658229351044,
      "learning_rate": 0.00012958061235012706,
      "loss": 0.2118,
      "step": 75
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.03474392369389534,
      "learning_rate": 0.0001273662990072083,
      "loss": 0.1465,
      "step": 76
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.033331647515296936,
      "learning_rate": 0.0001251373834511103,
      "loss": 0.2207,
      "step": 77
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.030515702441334724,
      "learning_rate": 0.0001228950549950134,
      "loss": 0.2121,
      "step": 78
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.19000689685344696,
      "eval_runtime": 153.8331,
      "eval_samples_per_second": 1.463,
      "eval_steps_per_second": 0.371,
      "step": 78
    },
    {
      "epoch": 1.01,
      "grad_norm": 0.03746604546904564,
      "learning_rate": 0.00012064051010900397,
      "loss": 0.1755,
      "step": 79
    },
    {
      "epoch": 1.02,
      "grad_norm": 0.032529812306165695,
      "learning_rate": 0.00011837495178165706,
      "loss": 0.2279,
      "step": 80
    },
    {
      "epoch": 1.01,
      "grad_norm": 0.02698327600955963,
      "learning_rate": 0.00011609958887814129,
      "loss": 0.1718,
      "step": 81
    },
    {
      "epoch": 1.02,
      "grad_norm": 0.030606260523200035,
      "learning_rate": 0.00011381563549518823,
      "loss": 0.2253,
      "step": 82
    },
    {
      "epoch": 1.04,
      "grad_norm": 0.03295598179101944,
      "learning_rate": 0.00011152431031326978,
      "loss": 0.1592,
      "step": 83
    },
    {
      "epoch": 1.05,
      "grad_norm": 0.04062403365969658,
      "learning_rate": 0.00010922683594633021,
      "loss": 0.2458,
      "step": 84
    },
    {
      "epoch": 1.06,
      "grad_norm": 0.04118546470999718,
      "learning_rate": 0.00010692443828941918,
      "loss": 0.1938,
      "step": 85
    },
    {
      "epoch": 1.07,
      "grad_norm": 0.03897158056497574,
      "learning_rate": 0.00010461834586457398,
      "loss": 0.1816,
      "step": 86
    },
    {
      "epoch": 1.09,
      "grad_norm": 0.031059524044394493,
      "learning_rate": 0.00010230978916530012,
      "loss": 0.2024,
      "step": 87
    },
    {
      "epoch": 1.1,
      "grad_norm": 0.0294091384857893,
      "learning_rate": 0.0001,
      "loss": 0.199,
      "step": 88
    },
    {
      "epoch": 1.11,
      "grad_norm": 0.03456791117787361,
      "learning_rate": 9.76902108346999e-05,
      "loss": 0.2616,
      "step": 89
    },
    {
      "epoch": 1.12,
      "grad_norm": 0.03276889771223068,
      "learning_rate": 9.538165413542607e-05,
      "loss": 0.2115,
      "step": 90
    },
    {
      "epoch": 1.14,
      "grad_norm": 0.03038157895207405,
      "learning_rate": 9.307556171058085e-05,
      "loss": 0.1721,
      "step": 91
    },
    {
      "epoch": 1.15,
      "grad_norm": 0.02885863371193409,
      "learning_rate": 9.077316405366981e-05,
      "loss": 0.2199,
      "step": 92
    },
    {
      "epoch": 1.16,
      "grad_norm": 0.03282390534877777,
      "learning_rate": 8.847568968673026e-05,
      "loss": 0.1885,
      "step": 93
    },
    {
      "epoch": 1.18,
      "grad_norm": 0.03559669107198715,
      "learning_rate": 8.61843645048118e-05,
      "loss": 0.2253,
      "step": 94
    },
    {
      "epoch": 1.19,
      "grad_norm": 0.027744270861148834,
      "learning_rate": 8.39004111218587e-05,
      "loss": 0.1825,
      "step": 95
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.035874996334314346,
      "learning_rate": 8.162504821834295e-05,
      "loss": 0.1983,
      "step": 96
    },
    {
      "epoch": 1.21,
      "grad_norm": 0.03325138986110687,
      "learning_rate": 7.935948989099605e-05,
      "loss": 0.1812,
      "step": 97
    },
    {
      "epoch": 1.23,
      "grad_norm": 0.028679154813289642,
      "learning_rate": 7.710494500498662e-05,
      "loss": 0.1875,
      "step": 98
    },
    {
      "epoch": 1.24,
      "grad_norm": 0.036103084683418274,
      "learning_rate": 7.486261654888973e-05,
      "loss": 0.2025,
      "step": 99
    },
    {
      "epoch": 1.25,
      "grad_norm": 0.0321945995092392,
      "learning_rate": 7.263370099279172e-05,
      "loss": 0.2342,
      "step": 100
    },
    {
      "epoch": 1.27,
      "grad_norm": 0.0304876621812582,
      "learning_rate": 7.041938764987297e-05,
      "loss": 0.1642,
      "step": 101
    },
    {
      "epoch": 1.28,
      "grad_norm": 0.03619467467069626,
      "learning_rate": 6.822085804180984e-05,
      "loss": 0.187,
      "step": 102
    },
    {
      "epoch": 1.29,
      "grad_norm": 0.03597356379032135,
      "learning_rate": 6.603928526833387e-05,
      "loss": 0.1343,
      "step": 103
    },
    {
      "epoch": 1.3,
      "grad_norm": 0.036540765315294266,
      "learning_rate": 6.387583338128471e-05,
      "loss": 0.2062,
      "step": 104
    },
    {
      "epoch": 1.32,
      "grad_norm": 0.03327038511633873,
      "learning_rate": 6.173165676349103e-05,
      "loss": 0.22,
      "step": 105
    },
    {
      "epoch": 1.33,
      "grad_norm": 0.03374246507883072,
      "learning_rate": 5.960789951281052e-05,
      "loss": 0.1765,
      "step": 106
    },
    {
      "epoch": 1.34,
      "grad_norm": 0.032783444970846176,
      "learning_rate": 5.750569483165784e-05,
      "loss": 0.2397,
      "step": 107
    },
    {
      "epoch": 1.35,
      "grad_norm": 0.03667570650577545,
      "learning_rate": 5.542616442234618e-05,
      "loss": 0.1841,
      "step": 108
    },
    {
      "epoch": 1.37,
      "grad_norm": 0.029467754065990448,
      "learning_rate": 5.337041788856518e-05,
      "loss": 0.2282,
      "step": 109
    },
    {
      "epoch": 1.38,
      "grad_norm": 0.03702455386519432,
      "learning_rate": 5.1339552143314384e-05,
      "loss": 0.2601,
      "step": 110
    },
    {
      "epoch": 1.39,
      "grad_norm": 0.033595018088817596,
      "learning_rate": 4.933465082360807e-05,
      "loss": 0.2077,
      "step": 111
    },
    {
      "epoch": 1.41,
      "grad_norm": 0.03349510580301285,
      "learning_rate": 4.735678371226441e-05,
      "loss": 0.1686,
      "step": 112
    },
    {
      "epoch": 1.42,
      "grad_norm": 0.03149344027042389,
      "learning_rate": 4.540700616708658e-05,
      "loss": 0.2075,
      "step": 113
    },
    {
      "epoch": 1.43,
      "grad_norm": 0.035017482936382294,
      "learning_rate": 4.3486358557740814e-05,
      "loss": 0.1462,
      "step": 114
    },
    {
      "epoch": 1.44,
      "grad_norm": 0.029670318588614464,
      "learning_rate": 4.159586571063236e-05,
      "loss": 0.1602,
      "step": 115
    },
    {
      "epoch": 1.46,
      "grad_norm": 0.037266362458467484,
      "learning_rate": 3.973653636207437e-05,
      "loss": 0.1798,
      "step": 116
    },
    {
      "epoch": 1.47,
      "grad_norm": 0.03251258283853531,
      "learning_rate": 3.7909362620042865e-05,
      "loss": 0.2512,
      "step": 117
    },
    {
      "epoch": 1.47,
      "eval_loss": 0.18417498469352722,
      "eval_runtime": 155.2041,
      "eval_samples_per_second": 1.45,
      "eval_steps_per_second": 0.367,
      "step": 117
    }
  ],
  "logging_steps": 1,
  "max_steps": 156,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 39,
  "total_flos": 1.5385293522940723e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}