{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.0,
  "eval_steps": 18,
  "global_step": 92,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.043478260869565216,
      "grad_norm": 39.07209396362305,
      "learning_rate": 0.0,
      "loss": 12.15625,
      "step": 1
    },
    {
      "epoch": 0.08695652173913043,
      "grad_norm": 37.47832107543945,
      "learning_rate": 1e-05,
      "loss": 11.5390625,
      "step": 2
    },
    {
      "epoch": 0.13043478260869565,
      "grad_norm": 34.37069320678711,
      "learning_rate": 2e-05,
      "loss": 10.71875,
      "step": 3
    },
    {
      "epoch": 0.17391304347826086,
      "grad_norm": 24.194438934326172,
      "learning_rate": 3e-05,
      "loss": 7.98046875,
      "step": 4
    },
    {
      "epoch": 0.21739130434782608,
      "grad_norm": 14.068402290344238,
      "learning_rate": 4e-05,
      "loss": 3.958984375,
      "step": 5
    },
    {
      "epoch": 0.2608695652173913,
      "grad_norm": 9.06344223022461,
      "learning_rate": 5e-05,
      "loss": 2.19873046875,
      "step": 6
    },
    {
      "epoch": 0.30434782608695654,
      "grad_norm": 3.2053747177124023,
      "learning_rate": 4.9985332146267735e-05,
      "loss": 0.994873046875,
      "step": 7
    },
    {
      "epoch": 0.34782608695652173,
      "grad_norm": 3.3570635318756104,
      "learning_rate": 4.994134770915388e-05,
      "loss": 0.752288818359375,
      "step": 8
    },
    {
      "epoch": 0.391304347826087,
      "grad_norm": 2.7033090591430664,
      "learning_rate": 4.98681040359731e-05,
      "loss": 0.448974609375,
      "step": 9
    },
    {
      "epoch": 0.43478260869565216,
      "grad_norm": 1.9760388135910034,
      "learning_rate": 4.9765696622501846e-05,
      "loss": 0.547119140625,
      "step": 10
    },
    {
      "epoch": 0.4782608695652174,
      "grad_norm": 1.2799606323242188,
      "learning_rate": 4.963425898847006e-05,
      "loss": 0.3215370178222656,
      "step": 11
    },
    {
      "epoch": 0.5217391304347826,
      "grad_norm": 1.4340665340423584,
      "learning_rate": 4.947396250347695e-05,
      "loss": 0.594482421875,
      "step": 12
    },
    {
      "epoch": 0.5652173913043478,
      "grad_norm": 1.2824798822402954,
      "learning_rate": 4.928501616355768e-05,
      "loss": 0.3077392578125,
      "step": 13
    },
    {
      "epoch": 0.6086956521739131,
      "grad_norm": 0.7614466547966003,
      "learning_rate": 4.906766631869243e-05,
      "loss": 0.244140625,
      "step": 14
    },
    {
      "epoch": 0.6521739130434783,
      "grad_norm": 0.7498453855514526,
      "learning_rate": 4.882219635161306e-05,
      "loss": 0.2913818359375,
      "step": 15
    },
    {
      "epoch": 0.6956521739130435,
      "grad_norm": 0.5524697303771973,
      "learning_rate": 4.854892630832603e-05,
      "loss": 0.28388214111328125,
      "step": 16
    },
    {
      "epoch": 0.7391304347826086,
      "grad_norm": 0.7023299932479858,
      "learning_rate": 4.82482124808335e-05,
      "loss": 0.28436279296875,
      "step": 17
    },
    {
      "epoch": 0.782608695652174,
      "grad_norm": 0.4874381721019745,
      "learning_rate": 4.7920446942596535e-05,
      "loss": 0.32769775390625,
      "step": 18
    },
    {
      "epoch": 0.8260869565217391,
      "grad_norm": 0.5234342813491821,
      "learning_rate": 4.7566057037346104e-05,
      "loss": 0.2855224609375,
      "step": 19
    },
    {
      "epoch": 0.8695652173913043,
      "grad_norm": 0.5836489796638489,
      "learning_rate": 4.718550482190837e-05,
      "loss": 0.231414794921875,
      "step": 20
    },
    {
      "epoch": 0.8695652173913043,
      "eval_loss": 0.2960205078125,
      "eval_runtime": 5.8443,
      "eval_samples_per_second": 0.513,
      "eval_steps_per_second": 0.171,
      "step": 20
    },
    {
      "epoch": 0.9130434782608695,
      "grad_norm": 0.520865797996521,
      "learning_rate": 4.677928646377076e-05,
      "loss": 0.2606201171875,
      "step": 21
    },
    {
      "epoch": 0.9565217391304348,
      "grad_norm": 0.5242786407470703,
      "learning_rate": 4.634793159417421e-05,
      "loss": 0.314208984375,
      "step": 22
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.4516263008117676,
      "learning_rate": 4.589200261757506e-05,
      "loss": 0.267333984375,
      "step": 23
    },
    {
      "epoch": 1.0434782608695652,
      "grad_norm": 0.42359301447868347,
      "learning_rate": 4.5412093978376986e-05,
      "loss": 0.224639892578125,
      "step": 24
    },
    {
      "epoch": 1.0869565217391304,
      "grad_norm": 0.36964306235313416,
      "learning_rate": 4.490883138588882e-05,
      "loss": 0.26383209228515625,
      "step": 25
    },
    {
      "epoch": 1.1304347826086956,
      "grad_norm": 0.6183728575706482,
      "learning_rate": 4.438287099851905e-05,
      "loss": 0.228271484375,
      "step": 26
    },
    {
      "epoch": 1.1739130434782608,
      "grad_norm": 0.7323199510574341,
      "learning_rate": 4.3834898568270444e-05,
      "loss": 0.22613072395324707,
      "step": 27
    },
    {
      "epoch": 1.2173913043478262,
      "grad_norm": 0.5516233444213867,
      "learning_rate": 4.326562854665021e-05,
      "loss": 0.14371299743652344,
      "step": 28
    },
    {
      "epoch": 1.2608695652173914,
      "grad_norm": 0.90989750623703,
      "learning_rate": 4.2675803153161565e-05,
      "loss": 0.373046875,
      "step": 29
    },
    {
      "epoch": 1.3043478260869565,
      "grad_norm": 0.7089672088623047,
      "learning_rate": 4.2066191407591125e-05,
      "loss": 0.20618438720703125,
      "step": 30
    },
    {
      "epoch": 1.3478260869565217,
      "grad_norm": 0.9005956649780273,
      "learning_rate": 4.1437588127353774e-05,
      "loss": 0.22583389282226562,
      "step": 31
    },
    {
      "epoch": 1.391304347826087,
      "grad_norm": 0.7037151455879211,
      "learning_rate": 4.07908128912024e-05,
      "loss": 0.2068347930908203,
      "step": 32
    },
    {
      "epoch": 1.434782608695652,
      "grad_norm": 0.5921205282211304,
      "learning_rate": 4.012670897065361e-05,
      "loss": 0.18936920166015625,
      "step": 33
    },
    {
      "epoch": 1.4782608695652173,
      "grad_norm": 0.31155166029930115,
      "learning_rate": 3.944614223052245e-05,
      "loss": 0.1267108917236328,
      "step": 34
    },
    {
      "epoch": 1.5217391304347827,
      "grad_norm": 0.7871601581573486,
      "learning_rate": 3.875e-05,
      "loss": 0.25548815727233887,
      "step": 35
    },
    {
      "epoch": 1.5652173913043477,
      "grad_norm": 1.0059878826141357,
      "learning_rate": 3.803918991574528e-05,
      "loss": 0.16361236572265625,
      "step": 36
    },
    {
      "epoch": 1.608695652173913,
      "grad_norm": 0.8261852860450745,
      "learning_rate": 3.7314638738500265e-05,
      "loss": 0.11863207817077637,
      "step": 37
    },
    {
      "epoch": 1.6521739130434783,
      "grad_norm": 0.5391600728034973,
      "learning_rate": 3.65772911447707e-05,
      "loss": 0.172882080078125,
      "step": 38
    },
    {
      "epoch": 1.6521739130434783,
      "eval_loss": 0.21588134765625,
      "eval_runtime": 4.9269,
      "eval_samples_per_second": 0.609,
      "eval_steps_per_second": 0.203,
      "step": 38
    },
    {
      "epoch": 1.6956521739130435,
      "grad_norm": 0.7111477851867676,
      "learning_rate": 3.582810849514808e-05,
      "loss": 0.16753196716308594,
      "step": 39
    },
    {
      "epoch": 1.7391304347826086,
      "grad_norm": 0.4621718227863312,
      "learning_rate": 3.506806758087894e-05,
      "loss": 0.14367055892944336,
      "step": 40
    },
    {
      "epoch": 1.7826086956521738,
      "grad_norm": 0.8741061091423035,
      "learning_rate": 3.429815935031538e-05,
      "loss": 0.16303253173828125,
      "step": 41
    },
    {
      "epoch": 1.8260869565217392,
      "grad_norm": 1.102018117904663,
      "learning_rate": 3.351938761690748e-05,
      "loss": 0.205902099609375,
      "step": 42
    },
    {
      "epoch": 1.8695652173913042,
      "grad_norm": 0.7496725916862488,
      "learning_rate": 3.273276775042199e-05,
      "loss": 0.14171886444091797,
      "step": 43
    },
    {
      "epoch": 1.9130434782608696,
      "grad_norm": 0.7663282155990601,
      "learning_rate": 3.1939325353094e-05,
      "loss": 0.14968395233154297,
      "step": 44
    },
    {
      "epoch": 1.9565217391304348,
      "grad_norm": 0.3884790241718292,
      "learning_rate": 3.114009492243721e-05,
      "loss": 0.15384674072265625,
      "step": 45
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.5156270861625671,
      "learning_rate": 3.033611850245651e-05,
      "loss": 0.1345977783203125,
      "step": 46
    },
    {
      "epoch": 2.0434782608695654,
      "grad_norm": 0.5794659852981567,
      "learning_rate": 2.9528444325021477e-05,
      "loss": 0.12321662902832031,
      "step": 47
    },
    {
      "epoch": 2.0869565217391304,
      "grad_norm": 0.36416205763816833,
      "learning_rate": 2.87181254431719e-05,
      "loss": 0.14786148071289062,
      "step": 48
    },
    {
      "epoch": 2.130434782608696,
      "grad_norm": 0.5249211192131042,
      "learning_rate": 2.790621835813761e-05,
      "loss": 0.1492919921875,
      "step": 49
    },
    {
      "epoch": 2.1739130434782608,
      "grad_norm": 0.4051743745803833,
      "learning_rate": 2.7093781641862387e-05,
      "loss": 0.0952908992767334,
      "step": 50
    },
    {
      "epoch": 2.217391304347826,
      "grad_norm": 0.26149407029151917,
      "learning_rate": 2.6281874556828108e-05,
      "loss": 0.08324813842773438,
      "step": 51
    },
    {
      "epoch": 2.260869565217391,
      "grad_norm": 0.5549029111862183,
      "learning_rate": 2.547155567497854e-05,
      "loss": 0.2025146484375,
      "step": 52
    },
    {
      "epoch": 2.3043478260869565,
      "grad_norm": 0.3528760075569153,
      "learning_rate": 2.4663881497543495e-05,
      "loss": 0.14881134033203125,
      "step": 53
    },
    {
      "epoch": 2.3478260869565215,
      "grad_norm": 0.5280032753944397,
      "learning_rate": 2.3859905077562796e-05,
      "loss": 0.176971435546875,
      "step": 54
    },
    {
      "epoch": 2.391304347826087,
      "grad_norm": 0.2940866947174072,
      "learning_rate": 2.3060674646906004e-05,
      "loss": 0.11214065551757812,
      "step": 55
    },
    {
      "epoch": 2.4347826086956523,
      "grad_norm": 0.46992143988609314,
      "learning_rate": 2.2267232249578024e-05,
      "loss": 0.1177825927734375,
      "step": 56
    },
    {
      "epoch": 2.4347826086956523,
      "eval_loss": 0.20440673828125,
      "eval_runtime": 4.915,
      "eval_samples_per_second": 0.61,
      "eval_steps_per_second": 0.203,
      "step": 56
    },
    {
      "epoch": 2.4782608695652173,
      "grad_norm": 0.3323771357536316,
      "learning_rate": 2.1480612383092536e-05,
      "loss": 0.09143924713134766,
      "step": 57
    },
    {
      "epoch": 2.5217391304347827,
      "grad_norm": 0.46137872338294983,
      "learning_rate": 2.0701840649684613e-05,
      "loss": 0.15516090393066406,
      "step": 58
    },
    {
      "epoch": 2.5652173913043477,
      "grad_norm": 0.26286780834198,
      "learning_rate": 1.993193241912106e-05,
      "loss": 0.06685352325439453,
      "step": 59
    },
    {
      "epoch": 2.608695652173913,
      "grad_norm": 0.2798925042152405,
      "learning_rate": 1.9171891504851925e-05,
      "loss": 0.0812065601348877,
      "step": 60
    },
    {
      "epoch": 2.6521739130434785,
      "grad_norm": 0.40160560607910156,
      "learning_rate": 1.842270885522931e-05,
      "loss": 0.07373046875,
      "step": 61
    },
    {
      "epoch": 2.6956521739130435,
      "grad_norm": 0.4648759067058563,
      "learning_rate": 1.7685361261499733e-05,
      "loss": 0.08234977722167969,
      "step": 62
    },
    {
      "epoch": 2.7391304347826084,
      "grad_norm": 0.22807049751281738,
      "learning_rate": 1.6960810084254726e-05,
      "loss": 0.0868840217590332,
      "step": 63
    },
    {
      "epoch": 2.782608695652174,
      "grad_norm": 0.3399467468261719,
      "learning_rate": 1.6250000000000005e-05,
      "loss": 0.083984375,
      "step": 64
    },
    {
      "epoch": 2.8260869565217392,
      "grad_norm": 0.5082715153694153,
      "learning_rate": 1.5553857769477553e-05,
      "loss": 0.10441970825195312,
      "step": 65
    },
    {
      "epoch": 2.869565217391304,
      "grad_norm": 0.47074776887893677,
      "learning_rate": 1.48732910293464e-05,
      "loss": 0.07005095481872559,
      "step": 66
    },
    {
      "epoch": 2.9130434782608696,
      "grad_norm": 0.7794228196144104,
      "learning_rate": 1.4209187108797607e-05,
      "loss": 0.10682344436645508,
      "step": 67
    },
    {
      "epoch": 2.9565217391304346,
      "grad_norm": 0.260376900434494,
      "learning_rate": 1.3562411872646235e-05,
      "loss": 0.09476900100708008,
      "step": 68
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.21531462669372559,
      "learning_rate": 1.293380859240888e-05,
      "loss": 0.047382354736328125,
      "step": 69
    },
    {
      "epoch": 3.0434782608695654,
      "grad_norm": 0.2740839123725891,
      "learning_rate": 1.232419684683844e-05,
      "loss": 0.0454249382019043,
      "step": 70
    },
    {
      "epoch": 3.0869565217391304,
      "grad_norm": 0.37194937467575073,
      "learning_rate": 1.1734371453349799e-05,
      "loss": 0.0704050064086914,
      "step": 71
    },
    {
      "epoch": 3.130434782608696,
      "grad_norm": 0.3591350317001343,
      "learning_rate": 1.1165101431729561e-05,
      "loss": 0.08895111083984375,
      "step": 72
    },
    {
      "epoch": 3.1739130434782608,
      "grad_norm": 0.21049781143665314,
      "learning_rate": 1.0617129001480949e-05,
      "loss": 0.044961631298065186,
      "step": 73
    },
    {
      "epoch": 3.217391304347826,
      "grad_norm": 0.11901774257421494,
      "learning_rate": 1.0091168614111182e-05,
      "loss": 0.01461946964263916,
      "step": 74
    },
    {
      "epoch": 3.217391304347826,
      "eval_loss": 0.19732666015625,
      "eval_runtime": 4.9262,
      "eval_samples_per_second": 0.609,
      "eval_steps_per_second": 0.203,
      "step": 74
    },
    {
      "epoch": 3.260869565217391,
      "grad_norm": 0.5734399557113647,
      "learning_rate": 9.587906021623016e-06,
      "loss": 0.12098979949951172,
      "step": 75
    },
    {
      "epoch": 3.3043478260869565,
      "grad_norm": 0.5726773738861084,
      "learning_rate": 9.107997382424935e-06,
      "loss": 0.06650257110595703,
      "step": 76
    },
    {
      "epoch": 3.3478260869565215,
      "grad_norm": 0.5507848858833313,
      "learning_rate": 8.652068405825798e-06,
      "loss": 0.11254596710205078,
      "step": 77
    },
    {
      "epoch": 3.391304347826087,
      "grad_norm": 0.23698930442333221,
      "learning_rate": 8.220713536229247e-06,
      "loss": 0.07490479946136475,
      "step": 78
    },
    {
      "epoch": 3.4347826086956523,
      "grad_norm": 0.3934122622013092,
      "learning_rate": 7.814495178091634e-06,
      "loss": 0.05574750900268555,
      "step": 79
    },
    {
      "epoch": 3.4782608695652173,
      "grad_norm": 0.19300000369548798,
      "learning_rate": 7.4339429626539e-06,
      "loss": 0.06278634071350098,
      "step": 80
    },
    {
      "epoch": 3.5217391304347827,
      "grad_norm": 0.3946453332901001,
      "learning_rate": 7.079553057403471e-06,
      "loss": 0.09804105758666992,
      "step": 81
    },
    {
      "epoch": 3.5652173913043477,
      "grad_norm": 0.2676098644733429,
      "learning_rate": 6.751787519166505e-06,
      "loss": 0.045120954513549805,
      "step": 82
    },
    {
      "epoch": 3.608695652173913,
      "grad_norm": 0.25669530034065247,
      "learning_rate": 6.451073691673979e-06,
      "loss": 0.04766273498535156,
      "step": 83
    },
    {
      "epoch": 3.6521739130434785,
      "grad_norm": 0.1960737109184265,
      "learning_rate": 6.177803648386948e-06,
      "loss": 0.030242919921875,
      "step": 84
    },
    {
      "epoch": 3.6956521739130435,
      "grad_norm": 0.35025930404663086,
      "learning_rate": 5.932333681307571e-06,
      "loss": 0.04735755920410156,
      "step": 85
    },
    {
      "epoch": 3.7391304347826084,
      "grad_norm": 0.2209789752960205,
      "learning_rate": 5.714983836442326e-06,
      "loss": 0.061261892318725586,
      "step": 86
    },
    {
      "epoch": 3.782608695652174,
      "grad_norm": 0.24673958122730255,
      "learning_rate": 5.526037496523051e-06,
      "loss": 0.05185413360595703,
      "step": 87
    },
    {
      "epoch": 3.8260869565217392,
      "grad_norm": 0.28059279918670654,
      "learning_rate": 5.36574101152994e-06,
      "loss": 0.0604095458984375,
      "step": 88
    },
    {
      "epoch": 3.869565217391304,
      "grad_norm": 0.6097232103347778,
      "learning_rate": 5.234303377498163e-06,
      "loss": 0.0322948694229126,
      "step": 89
    },
    {
      "epoch": 3.9130434782608696,
      "grad_norm": 0.42037317156791687,
      "learning_rate": 5.1318959640269095e-06,
      "loss": 0.0640861988067627,
      "step": 90
    },
    {
      "epoch": 3.9565217391304346,
      "grad_norm": 0.22852636873722076,
      "learning_rate": 5.058652290846131e-06,
      "loss": 0.07720410823822021,
      "step": 91
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.13465619087219238,
      "learning_rate": 5.014667853732269e-06,
      "loss": 0.02185821533203125,
      "step": 92
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.2081298828125,
      "eval_runtime": 4.9238,
      "eval_samples_per_second": 0.609,
      "eval_steps_per_second": 0.203,
      "step": 92
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 92,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 0,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5.0147843800825856e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}