{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.7392996108949417,
  "eval_steps": 8,
  "global_step": 88,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0311284046692607,
      "grad_norm": 0.5707955360412598,
      "learning_rate": 1e-05,
      "loss": 1.8935,
      "step": 1
    },
    {
      "epoch": 0.0311284046692607,
      "eval_loss": 1.8884226083755493,
      "eval_runtime": 34.2567,
      "eval_samples_per_second": 29.104,
      "eval_steps_per_second": 0.321,
      "step": 1
    },
    {
      "epoch": 0.0622568093385214,
      "grad_norm": 0.5781293511390686,
      "learning_rate": 2e-05,
      "loss": 1.879,
      "step": 2
    },
    {
      "epoch": 0.0933852140077821,
      "grad_norm": 0.5720934271812439,
      "learning_rate": 3e-05,
      "loss": 1.8848,
      "step": 3
    },
    {
      "epoch": 0.1245136186770428,
      "grad_norm": 0.580179750919342,
      "learning_rate": 4e-05,
      "loss": 1.8845,
      "step": 4
    },
    {
      "epoch": 0.1556420233463035,
      "grad_norm": 0.6264262795448303,
      "learning_rate": 5e-05,
      "loss": 1.8758,
      "step": 5
    },
    {
      "epoch": 0.1867704280155642,
      "grad_norm": 0.643973708152771,
      "learning_rate": 6e-05,
      "loss": 1.8309,
      "step": 6
    },
    {
      "epoch": 0.2178988326848249,
      "grad_norm": 0.6367993354797363,
      "learning_rate": 7e-05,
      "loss": 1.7743,
      "step": 7
    },
    {
      "epoch": 0.2490272373540856,
      "grad_norm": 0.5833392143249512,
      "learning_rate": 8e-05,
      "loss": 1.6965,
      "step": 8
    },
    {
      "epoch": 0.2490272373540856,
      "eval_loss": 1.5970573425292969,
      "eval_runtime": 34.32,
      "eval_samples_per_second": 29.05,
      "eval_steps_per_second": 0.321,
      "step": 8
    },
    {
      "epoch": 0.2801556420233463,
      "grad_norm": 0.5133880972862244,
      "learning_rate": 9e-05,
      "loss": 1.5915,
      "step": 9
    },
    {
      "epoch": 0.311284046692607,
      "grad_norm": 0.42409589886665344,
      "learning_rate": 0.0001,
      "loss": 1.5128,
      "step": 10
    },
    {
      "epoch": 0.3424124513618677,
      "grad_norm": 0.3264746069908142,
      "learning_rate": 0.00011000000000000002,
      "loss": 1.4567,
      "step": 11
    },
    {
      "epoch": 0.3735408560311284,
      "grad_norm": 0.2589164078235626,
      "learning_rate": 0.00012,
      "loss": 1.4249,
      "step": 12
    },
    {
      "epoch": 0.4046692607003891,
      "grad_norm": 0.3931436538696289,
      "learning_rate": 0.00013000000000000002,
      "loss": 1.4125,
      "step": 13
    },
    {
      "epoch": 0.4357976653696498,
      "grad_norm": 0.5455179810523987,
      "learning_rate": 0.00014,
      "loss": 1.4079,
      "step": 14
    },
    {
      "epoch": 0.4669260700389105,
      "grad_norm": 0.5418187379837036,
      "learning_rate": 0.00015000000000000001,
      "loss": 1.4031,
      "step": 15
    },
    {
      "epoch": 0.4980544747081712,
      "grad_norm": 0.42387455701828003,
      "learning_rate": 0.00016,
      "loss": 1.3733,
      "step": 16
    },
    {
      "epoch": 0.4980544747081712,
      "eval_loss": 1.3445571660995483,
      "eval_runtime": 34.3152,
      "eval_samples_per_second": 29.054,
      "eval_steps_per_second": 0.321,
      "step": 16
    },
    {
      "epoch": 0.5291828793774319,
      "grad_norm": 0.2986687123775482,
      "learning_rate": 0.00017,
      "loss": 1.3505,
      "step": 17
    },
    {
      "epoch": 0.5603112840466926,
      "grad_norm": 0.2099975347518921,
      "learning_rate": 0.00018,
      "loss": 1.3243,
      "step": 18
    },
    {
      "epoch": 0.5914396887159533,
      "grad_norm": 0.16759291291236877,
      "learning_rate": 0.00019,
      "loss": 1.3056,
      "step": 19
    },
    {
      "epoch": 0.622568093385214,
      "grad_norm": 0.16132138669490814,
      "learning_rate": 0.0002,
      "loss": 1.3014,
      "step": 20
    },
    {
      "epoch": 0.6536964980544747,
      "grad_norm": 0.17767557501792908,
      "learning_rate": 0.0001999145758387301,
      "loss": 1.2932,
      "step": 21
    },
    {
      "epoch": 0.6848249027237354,
      "grad_norm": 0.19573098421096802,
      "learning_rate": 0.000199658449300667,
      "loss": 1.2771,
      "step": 22
    },
    {
      "epoch": 0.7159533073929961,
      "grad_norm": 0.19915379583835602,
      "learning_rate": 0.0001992320579737045,
      "loss": 1.2762,
      "step": 23
    },
    {
      "epoch": 0.7470817120622568,
      "grad_norm": 0.17230945825576782,
      "learning_rate": 0.00019863613034027224,
      "loss": 1.2466,
      "step": 24
    },
    {
      "epoch": 0.7470817120622568,
      "eval_loss": 1.2462533712387085,
      "eval_runtime": 34.3129,
      "eval_samples_per_second": 29.056,
      "eval_steps_per_second": 0.321,
      "step": 24
    },
    {
      "epoch": 0.7782101167315175,
      "grad_norm": 0.13044685125350952,
      "learning_rate": 0.00019787168453273544,
      "loss": 1.2402,
      "step": 25
    },
    {
      "epoch": 0.8093385214007782,
      "grad_norm": 0.09282781183719635,
      "learning_rate": 0.00019694002659393305,
      "loss": 1.234,
      "step": 26
    },
    {
      "epoch": 0.8404669260700389,
      "grad_norm": 0.10575597733259201,
      "learning_rate": 0.0001958427482458253,
      "loss": 1.2214,
      "step": 27
    },
    {
      "epoch": 0.8715953307392996,
      "grad_norm": 0.14210504293441772,
      "learning_rate": 0.00019458172417006347,
      "loss": 1.2185,
      "step": 28
    },
    {
      "epoch": 0.9027237354085603,
      "grad_norm": 0.17919066548347473,
      "learning_rate": 0.0001931591088051279,
      "loss": 1.2025,
      "step": 29
    },
    {
      "epoch": 0.933852140077821,
      "grad_norm": 0.16358336806297302,
      "learning_rate": 0.00019157733266550575,
      "loss": 1.2032,
      "step": 30
    },
    {
      "epoch": 0.9649805447470817,
      "grad_norm": 0.13862887024879456,
      "learning_rate": 0.0001898390981891979,
      "loss": 1.197,
      "step": 31
    },
    {
      "epoch": 0.9961089494163424,
      "grad_norm": 0.11003394424915314,
      "learning_rate": 0.0001879473751206489,
      "loss": 1.1852,
      "step": 32
    },
    {
      "epoch": 0.9961089494163424,
      "eval_loss": 1.1821681261062622,
      "eval_runtime": 34.3117,
      "eval_samples_per_second": 29.057,
      "eval_steps_per_second": 0.321,
      "step": 32
    },
    {
      "epoch": 1.027237354085603,
      "grad_norm": 0.08200129121541977,
      "learning_rate": 0.00018590539543698854,
      "loss": 1.178,
      "step": 33
    },
    {
      "epoch": 1.0583657587548638,
      "grad_norm": 0.07455576211214066,
      "learning_rate": 0.00018371664782625287,
      "loss": 1.1725,
      "step": 34
    },
    {
      "epoch": 1.0894941634241244,
      "grad_norm": 0.08433058857917786,
      "learning_rate": 0.0001813848717270195,
      "loss": 1.1569,
      "step": 35
    },
    {
      "epoch": 1.1206225680933852,
      "grad_norm": 0.09246356040239334,
      "learning_rate": 0.00017891405093963938,
      "loss": 1.1627,
      "step": 36
    },
    {
      "epoch": 1.1517509727626458,
      "grad_norm": 0.09312273561954498,
      "learning_rate": 0.00017630840681998066,
      "loss": 1.1526,
      "step": 37
    },
    {
      "epoch": 1.1828793774319066,
      "grad_norm": 0.08373520523309708,
      "learning_rate": 0.00017357239106731317,
      "loss": 1.1456,
      "step": 38
    },
    {
      "epoch": 1.2140077821011672,
      "grad_norm": 0.07111110538244247,
      "learning_rate": 0.00017071067811865476,
      "loss": 1.1531,
      "step": 39
    },
    {
      "epoch": 1.245136186770428,
      "grad_norm": 0.06889671832323074,
      "learning_rate": 0.00016772815716257412,
      "loss": 1.1444,
      "step": 40
    },
    {
      "epoch": 1.245136186770428,
      "eval_loss": 1.1379262208938599,
      "eval_runtime": 34.3236,
      "eval_samples_per_second": 29.047,
      "eval_steps_per_second": 0.32,
      "step": 40
    },
    {
      "epoch": 1.2762645914396886,
      "grad_norm": 0.06582967936992645,
      "learning_rate": 0.00016462992378609407,
      "loss": 1.1335,
      "step": 41
    },
    {
      "epoch": 1.3073929961089494,
      "grad_norm": 0.07529184967279434,
      "learning_rate": 0.0001614212712689668,
      "loss": 1.1292,
      "step": 42
    },
    {
      "epoch": 1.3385214007782102,
      "grad_norm": 0.07816017419099808,
      "learning_rate": 0.00015810768154019385,
      "loss": 1.1293,
      "step": 43
    },
    {
      "epoch": 1.3696498054474708,
      "grad_norm": 0.08063483238220215,
      "learning_rate": 0.00015469481581224272,
      "loss": 1.1161,
      "step": 44
    },
    {
      "epoch": 1.4007782101167314,
      "grad_norm": 0.06947366893291473,
      "learning_rate": 0.00015118850490896012,
      "loss": 1.1168,
      "step": 45
    },
    {
      "epoch": 1.4319066147859922,
      "grad_norm": 0.05603436380624771,
      "learning_rate": 0.00014759473930370736,
      "loss": 1.1147,
      "step": 46
    },
    {
      "epoch": 1.463035019455253,
      "grad_norm": 0.055858004838228226,
      "learning_rate": 0.00014391965888473703,
      "loss": 1.1123,
      "step": 47
    },
    {
      "epoch": 1.4941634241245136,
      "grad_norm": 0.0600324422121048,
      "learning_rate": 0.00014016954246529696,
      "loss": 1.0986,
      "step": 48
    },
    {
      "epoch": 1.4941634241245136,
      "eval_loss": 1.1052128076553345,
      "eval_runtime": 34.2952,
      "eval_samples_per_second": 29.071,
      "eval_steps_per_second": 0.321,
      "step": 48
    },
    {
      "epoch": 1.5252918287937742,
      "grad_norm": 0.0596173070371151,
      "learning_rate": 0.00013635079705638298,
      "loss": 1.0949,
      "step": 49
    },
    {
      "epoch": 1.556420233463035,
      "grad_norm": 0.06981530040502548,
      "learning_rate": 0.00013246994692046836,
      "loss": 1.1,
      "step": 50
    },
    {
      "epoch": 1.5875486381322959,
      "grad_norm": 0.058555856347084045,
      "learning_rate": 0.00012853362242491053,
      "loss": 1.0946,
      "step": 51
    },
    {
      "epoch": 1.6186770428015564,
      "grad_norm": 0.052131447941064835,
      "learning_rate": 0.00012454854871407994,
      "loss": 1.096,
      "step": 52
    },
    {
      "epoch": 1.649805447470817,
      "grad_norm": 0.05138020217418671,
      "learning_rate": 0.00012052153421956342,
      "loss": 1.0948,
      "step": 53
    },
    {
      "epoch": 1.6809338521400778,
      "grad_norm": 0.055884215980768204,
      "learning_rate": 0.00011645945902807341,
      "loss": 1.0868,
      "step": 54
    },
    {
      "epoch": 1.7120622568093387,
      "grad_norm": 0.056635960936546326,
      "learning_rate": 0.00011236926312693479,
      "loss": 1.0782,
      "step": 55
    },
    {
      "epoch": 1.7431906614785992,
      "grad_norm": 0.05791952833533287,
      "learning_rate": 0.00010825793454723325,
      "loss": 1.0774,
      "step": 56
    },
    {
      "epoch": 1.7431906614785992,
      "eval_loss": 1.0816473960876465,
      "eval_runtime": 34.308,
      "eval_samples_per_second": 29.06,
      "eval_steps_per_second": 0.321,
      "step": 56
    },
    {
      "epoch": 1.7743190661478598,
      "grad_norm": 0.05655137449502945,
      "learning_rate": 0.00010413249742488131,
      "loss": 1.0793,
      "step": 57
    },
    {
      "epoch": 1.8054474708171206,
      "grad_norm": 0.05930772423744202,
      "learning_rate": 0.0001,
      "loss": 1.0765,
      "step": 58
    },
    {
      "epoch": 1.8365758754863815,
      "grad_norm": 0.056934159249067307,
      "learning_rate": 9.586750257511867e-05,
      "loss": 1.0825,
      "step": 59
    },
    {
      "epoch": 1.867704280155642,
      "grad_norm": 0.05056174844503403,
      "learning_rate": 9.174206545276677e-05,
      "loss": 1.074,
      "step": 60
    },
    {
      "epoch": 1.8988326848249026,
      "grad_norm": 0.05416735261678696,
      "learning_rate": 8.763073687306524e-05,
      "loss": 1.0731,
      "step": 61
    },
    {
      "epoch": 1.9299610894941635,
      "grad_norm": 0.05306009575724602,
      "learning_rate": 8.35405409719266e-05,
      "loss": 1.0646,
      "step": 62
    },
    {
      "epoch": 1.9610894941634243,
      "grad_norm": 0.054572440683841705,
      "learning_rate": 7.947846578043659e-05,
      "loss": 1.0697,
      "step": 63
    },
    {
      "epoch": 1.9922178988326849,
      "grad_norm": 0.051973506808280945,
      "learning_rate": 7.54514512859201e-05,
      "loss": 1.065,
      "step": 64
    },
    {
      "epoch": 1.9922178988326849,
      "eval_loss": 1.0657449960708618,
      "eval_runtime": 34.2892,
      "eval_samples_per_second": 29.076,
      "eval_steps_per_second": 0.321,
      "step": 64
    },
    {
      "epoch": 2.0233463035019454,
      "grad_norm": 0.048152584582567215,
      "learning_rate": 7.146637757508949e-05,
      "loss": 1.0629,
      "step": 65
    },
    {
      "epoch": 2.054474708171206,
      "grad_norm": 0.04994530603289604,
      "learning_rate": 6.753005307953167e-05,
      "loss": 1.0516,
      "step": 66
    },
    {
      "epoch": 2.085603112840467,
      "grad_norm": 0.05009295791387558,
      "learning_rate": 6.3649202943617e-05,
      "loss": 1.0526,
      "step": 67
    },
    {
      "epoch": 2.1167315175097277,
      "grad_norm": 0.05345555767416954,
      "learning_rate": 5.983045753470308e-05,
      "loss": 1.0553,
      "step": 68
    },
    {
      "epoch": 2.1478599221789882,
      "grad_norm": 0.04756650701165199,
      "learning_rate": 5.608034111526298e-05,
      "loss": 1.059,
      "step": 69
    },
    {
      "epoch": 2.178988326848249,
      "grad_norm": 0.04925397038459778,
      "learning_rate": 5.240526069629265e-05,
      "loss": 1.0508,
      "step": 70
    },
    {
      "epoch": 2.21011673151751,
      "grad_norm": 0.05096421390771866,
      "learning_rate": 4.8811495091039926e-05,
      "loss": 1.0472,
      "step": 71
    },
    {
      "epoch": 2.2412451361867705,
      "grad_norm": 0.047330863773822784,
      "learning_rate": 4.530518418775733e-05,
      "loss": 1.055,
      "step": 72
    },
    {
      "epoch": 2.2412451361867705,
      "eval_loss": 1.0550851821899414,
      "eval_runtime": 34.2738,
      "eval_samples_per_second": 29.089,
      "eval_steps_per_second": 0.321,
      "step": 72
    },
    {
      "epoch": 2.272373540856031,
      "grad_norm": 0.04690932855010033,
      "learning_rate": 4.189231845980618e-05,
      "loss": 1.0495,
      "step": 73
    },
    {
      "epoch": 2.3035019455252916,
      "grad_norm": 0.04692551866173744,
      "learning_rate": 3.857872873103322e-05,
      "loss": 1.0561,
      "step": 74
    },
    {
      "epoch": 2.3346303501945527,
      "grad_norm": 0.04910856485366821,
      "learning_rate": 3.53700762139059e-05,
      "loss": 1.0459,
      "step": 75
    },
    {
      "epoch": 2.3657587548638133,
      "grad_norm": 0.04869484528899193,
      "learning_rate": 3.227184283742591e-05,
      "loss": 1.0373,
      "step": 76
    },
    {
      "epoch": 2.396887159533074,
      "grad_norm": 0.045992154628038406,
      "learning_rate": 2.9289321881345254e-05,
      "loss": 1.0306,
      "step": 77
    },
    {
      "epoch": 2.4280155642023344,
      "grad_norm": 0.04799241945147514,
      "learning_rate": 2.6427608932686843e-05,
      "loss": 1.051,
      "step": 78
    },
    {
      "epoch": 2.4591439688715955,
      "grad_norm": 0.04848311096429825,
      "learning_rate": 2.3691593180019366e-05,
      "loss": 1.0408,
      "step": 79
    },
    {
      "epoch": 2.490272373540856,
      "grad_norm": 0.04728139936923981,
      "learning_rate": 2.1085949060360654e-05,
      "loss": 1.0438,
      "step": 80
    },
    {
      "epoch": 2.490272373540856,
      "eval_loss": 1.0484414100646973,
      "eval_runtime": 34.2834,
      "eval_samples_per_second": 29.081,
      "eval_steps_per_second": 0.321,
      "step": 80
    },
    {
      "epoch": 2.5214007782101167,
      "grad_norm": 0.04541860893368721,
      "learning_rate": 1.861512827298051e-05,
      "loss": 1.0422,
      "step": 81
    },
    {
      "epoch": 2.5525291828793772,
      "grad_norm": 0.04615321755409241,
      "learning_rate": 1.6283352173747145e-05,
      "loss": 1.0388,
      "step": 82
    },
    {
      "epoch": 2.5836575875486383,
      "grad_norm": 0.04621463268995285,
      "learning_rate": 1.4094604563011472e-05,
      "loss": 1.0442,
      "step": 83
    },
    {
      "epoch": 2.614785992217899,
      "grad_norm": 0.045208945870399475,
      "learning_rate": 1.2052624879351104e-05,
      "loss": 1.0441,
      "step": 84
    },
    {
      "epoch": 2.6459143968871595,
      "grad_norm": 0.04617554694414139,
      "learning_rate": 1.0160901810802115e-05,
      "loss": 1.0395,
      "step": 85
    },
    {
      "epoch": 2.6770428015564205,
      "grad_norm": 0.043534088879823685,
      "learning_rate": 8.422667334494249e-06,
      "loss": 1.0463,
      "step": 86
    },
    {
      "epoch": 2.708171206225681,
      "grad_norm": 0.04501954838633537,
      "learning_rate": 6.840891194872112e-06,
      "loss": 1.0426,
      "step": 87
    },
    {
      "epoch": 2.7392996108949417,
      "grad_norm": 0.04564449191093445,
      "learning_rate": 5.418275829936537e-06,
      "loss": 1.0394,
      "step": 88
    },
    {
      "epoch": 2.7392996108949417,
      "eval_loss": 1.0463460683822632,
      "eval_runtime": 34.2795,
      "eval_samples_per_second": 29.084,
      "eval_steps_per_second": 0.321,
      "step": 88
    }
  ],
  "logging_steps": 1,
  "max_steps": 96,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 11,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.5197330643132875e+19,
  "train_batch_size": 12,
  "trial_name": null,
  "trial_params": null
}