{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "eval_steps": 100,
  "global_step": 5000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2,
      "grad_norm": 0.1518353819847107,
      "learning_rate": 5e-05,
      "loss": 1.1279,
      "step": 100
    },
    {
      "epoch": 0.2,
      "eval_loss": 1.0322937965393066,
      "eval_runtime": 186.6896,
      "eval_samples_per_second": 0.761,
      "eval_steps_per_second": 0.761,
      "step": 100
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.3351062834262848,
      "learning_rate": 4.89795918367347e-05,
      "loss": 1.0369,
      "step": 200
    },
    {
      "epoch": 0.4,
      "eval_loss": 0.9710885882377625,
      "eval_runtime": 182.1459,
      "eval_samples_per_second": 0.78,
      "eval_steps_per_second": 0.78,
      "step": 200
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.3858621120452881,
      "learning_rate": 4.795918367346939e-05,
      "loss": 1.004,
      "step": 300
    },
    {
      "epoch": 0.6,
      "eval_loss": 0.9358227252960205,
      "eval_runtime": 181.6505,
      "eval_samples_per_second": 0.782,
      "eval_steps_per_second": 0.782,
      "step": 300
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.40071049332618713,
      "learning_rate": 4.6938775510204086e-05,
      "loss": 0.955,
      "step": 400
    },
    {
      "epoch": 0.8,
      "eval_loss": 0.9108649492263794,
      "eval_runtime": 181.8409,
      "eval_samples_per_second": 0.781,
      "eval_steps_per_second": 0.781,
      "step": 400
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.41050222516059875,
      "learning_rate": 4.591836734693878e-05,
      "loss": 0.9817,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.8939587473869324,
      "eval_runtime": 181.68,
      "eval_samples_per_second": 0.782,
      "eval_steps_per_second": 0.782,
      "step": 500
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.5476531982421875,
      "learning_rate": 4.4897959183673474e-05,
      "loss": 0.8819,
      "step": 600
    },
    {
      "epoch": 1.2,
      "eval_loss": 0.8783901333808899,
      "eval_runtime": 182.0958,
      "eval_samples_per_second": 0.78,
      "eval_steps_per_second": 0.78,
      "step": 600
    },
    {
      "epoch": 1.4,
      "grad_norm": 0.5108929872512817,
      "learning_rate": 4.387755102040816e-05,
      "loss": 0.8373,
      "step": 700
    },
    {
      "epoch": 1.4,
      "eval_loss": 0.8652149438858032,
      "eval_runtime": 181.7715,
      "eval_samples_per_second": 0.781,
      "eval_steps_per_second": 0.781,
      "step": 700
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.5909234881401062,
      "learning_rate": 4.2857142857142856e-05,
      "loss": 0.8588,
      "step": 800
    },
    {
      "epoch": 1.6,
      "eval_loss": 0.8540464639663696,
      "eval_runtime": 181.5147,
      "eval_samples_per_second": 0.782,
      "eval_steps_per_second": 0.782,
      "step": 800
    },
    {
      "epoch": 1.8,
      "grad_norm": 0.5984646081924438,
      "learning_rate": 4.183673469387756e-05,
      "loss": 0.8656,
      "step": 900
    },
    {
      "epoch": 1.8,
      "eval_loss": 0.8421266674995422,
      "eval_runtime": 181.6623,
      "eval_samples_per_second": 0.782,
      "eval_steps_per_second": 0.782,
      "step": 900
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.6141483783721924,
      "learning_rate": 4.0816326530612245e-05,
      "loss": 0.8625,
      "step": 1000
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.8279427289962769,
      "eval_runtime": 181.7062,
      "eval_samples_per_second": 0.781,
      "eval_steps_per_second": 0.781,
      "step": 1000
    },
    {
      "epoch": 2.2,
      "grad_norm": 0.6167280077934265,
      "learning_rate": 3.979591836734694e-05,
      "loss": 0.8022,
      "step": 1100
    },
    {
      "epoch": 2.2,
      "eval_loss": 0.8208828568458557,
      "eval_runtime": 182.0493,
      "eval_samples_per_second": 0.78,
      "eval_steps_per_second": 0.78,
      "step": 1100
    },
    {
      "epoch": 2.4,
      "grad_norm": 0.6359805464744568,
      "learning_rate": 3.8775510204081634e-05,
      "loss": 0.7596,
      "step": 1200
    },
    {
      "epoch": 2.4,
      "eval_loss": 0.814860463142395,
      "eval_runtime": 181.8235,
      "eval_samples_per_second": 0.781,
      "eval_steps_per_second": 0.781,
      "step": 1200
    },
    {
      "epoch": 2.6,
      "grad_norm": 0.723619818687439,
      "learning_rate": 3.775510204081633e-05,
      "loss": 0.7715,
      "step": 1300
    },
    {
      "epoch": 2.6,
      "eval_loss": 0.7978885173797607,
      "eval_runtime": 182.0693,
      "eval_samples_per_second": 0.78,
      "eval_steps_per_second": 0.78,
      "step": 1300
    },
    {
      "epoch": 2.8,
      "grad_norm": 0.7059823274612427,
      "learning_rate": 3.673469387755102e-05,
      "loss": 0.7314,
      "step": 1400
    },
    {
      "epoch": 2.8,
      "eval_loss": 0.7922654747962952,
      "eval_runtime": 182.558,
      "eval_samples_per_second": 0.778,
      "eval_steps_per_second": 0.778,
      "step": 1400
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.5519229173660278,
      "learning_rate": 3.571428571428572e-05,
      "loss": 0.7402,
      "step": 1500
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.782792329788208,
      "eval_runtime": 182.1181,
      "eval_samples_per_second": 0.78,
      "eval_steps_per_second": 0.78,
      "step": 1500
    },
    {
      "epoch": 3.2,
      "grad_norm": 0.6199519634246826,
      "learning_rate": 3.469387755102041e-05,
      "loss": 0.6767,
      "step": 1600
    },
    {
      "epoch": 3.2,
      "eval_loss": 0.7840178608894348,
      "eval_runtime": 178.9485,
      "eval_samples_per_second": 0.794,
      "eval_steps_per_second": 0.794,
      "step": 1600
    },
    {
      "epoch": 3.4,
      "grad_norm": 0.642126202583313,
      "learning_rate": 3.36734693877551e-05,
      "loss": 0.6481,
      "step": 1700
    },
    {
      "epoch": 3.4,
      "eval_loss": 0.7741957306861877,
      "eval_runtime": 179.1517,
      "eval_samples_per_second": 0.793,
      "eval_steps_per_second": 0.793,
      "step": 1700
    },
    {
      "epoch": 3.6,
      "grad_norm": 0.7174199819564819,
      "learning_rate": 3.265306122448979e-05,
      "loss": 0.6689,
      "step": 1800
    },
    {
      "epoch": 3.6,
      "eval_loss": 0.7609220743179321,
      "eval_runtime": 180.4754,
      "eval_samples_per_second": 0.787,
      "eval_steps_per_second": 0.787,
      "step": 1800
    },
    {
      "epoch": 3.8,
      "grad_norm": 0.6541227698326111,
      "learning_rate": 3.1632653061224494e-05,
      "loss": 0.6652,
      "step": 1900
    },
    {
      "epoch": 3.8,
      "eval_loss": 0.75916987657547,
      "eval_runtime": 179.9759,
      "eval_samples_per_second": 0.789,
      "eval_steps_per_second": 0.789,
      "step": 1900
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.889131486415863,
      "learning_rate": 3.061224489795919e-05,
      "loss": 0.6938,
      "step": 2000
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.7402585744857788,
      "eval_runtime": 181.8785,
      "eval_samples_per_second": 0.781,
      "eval_steps_per_second": 0.781,
      "step": 2000
    },
    {
      "epoch": 4.2,
      "grad_norm": 0.9711707234382629,
      "learning_rate": 2.959183673469388e-05,
      "loss": 0.5935,
      "step": 2100
    },
    {
      "epoch": 4.2,
      "eval_loss": 0.7453898191452026,
      "eval_runtime": 181.9504,
      "eval_samples_per_second": 0.78,
      "eval_steps_per_second": 0.78,
      "step": 2100
    },
    {
      "epoch": 4.4,
      "grad_norm": 1.0045188665390015,
      "learning_rate": 2.857142857142857e-05,
      "loss": 0.5731,
      "step": 2200
    },
    {
      "epoch": 4.4,
      "eval_loss": 0.7436273097991943,
      "eval_runtime": 182.1074,
      "eval_samples_per_second": 0.78,
      "eval_steps_per_second": 0.78,
      "step": 2200
    },
    {
      "epoch": 4.6,
      "grad_norm": 0.7766691446304321,
      "learning_rate": 2.7551020408163265e-05,
      "loss": 0.6094,
      "step": 2300
    },
    {
      "epoch": 4.6,
      "eval_loss": 0.7243757843971252,
      "eval_runtime": 182.5282,
      "eval_samples_per_second": 0.778,
      "eval_steps_per_second": 0.778,
      "step": 2300
    },
    {
      "epoch": 4.8,
      "grad_norm": 0.8756657242774963,
      "learning_rate": 2.6530612244897963e-05,
      "loss": 0.5904,
      "step": 2400
    },
    {
      "epoch": 4.8,
      "eval_loss": 0.7280852794647217,
      "eval_runtime": 182.6182,
      "eval_samples_per_second": 0.778,
      "eval_steps_per_second": 0.778,
      "step": 2400
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.9650371670722961,
      "learning_rate": 2.5510204081632654e-05,
      "loss": 0.5649,
      "step": 2500
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.714751124382019,
      "eval_runtime": 182.1779,
      "eval_samples_per_second": 0.779,
      "eval_steps_per_second": 0.779,
      "step": 2500
    },
    {
      "epoch": 5.2,
      "grad_norm": 0.9919518828392029,
      "learning_rate": 2.448979591836735e-05,
      "loss": 0.5211,
      "step": 2600
    },
    {
      "epoch": 5.2,
      "eval_loss": 0.719607949256897,
      "eval_runtime": 182.2031,
      "eval_samples_per_second": 0.779,
      "eval_steps_per_second": 0.779,
      "step": 2600
    },
    {
      "epoch": 5.4,
      "grad_norm": 0.979714572429657,
      "learning_rate": 2.3469387755102043e-05,
      "loss": 0.536,
      "step": 2700
    },
    {
      "epoch": 5.4,
      "eval_loss": 0.7172472476959229,
      "eval_runtime": 182.2624,
      "eval_samples_per_second": 0.779,
      "eval_steps_per_second": 0.779,
      "step": 2700
    },
    {
      "epoch": 5.6,
      "grad_norm": 0.9355886578559875,
      "learning_rate": 2.2448979591836737e-05,
      "loss": 0.4957,
      "step": 2800
    },
    {
      "epoch": 5.6,
      "eval_loss": 0.7044922113418579,
      "eval_runtime": 182.1756,
      "eval_samples_per_second": 0.779,
      "eval_steps_per_second": 0.779,
      "step": 2800
    },
    {
      "epoch": 5.8,
      "grad_norm": 0.907577633857727,
      "learning_rate": 2.1428571428571428e-05,
      "loss": 0.4935,
      "step": 2900
    },
    {
      "epoch": 5.8,
      "eval_loss": 0.6986051201820374,
      "eval_runtime": 182.1701,
      "eval_samples_per_second": 0.779,
      "eval_steps_per_second": 0.779,
      "step": 2900
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.6769922971725464,
      "learning_rate": 2.0408163265306123e-05,
      "loss": 0.5038,
      "step": 3000
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.6920154690742493,
      "eval_runtime": 182.1079,
      "eval_samples_per_second": 0.78,
      "eval_steps_per_second": 0.78,
      "step": 3000
    },
    {
      "epoch": 6.2,
      "grad_norm": 0.7743176817893982,
      "learning_rate": 1.9387755102040817e-05,
      "loss": 0.4357,
      "step": 3100
    },
    {
      "epoch": 6.2,
      "eval_loss": 0.7011306285858154,
      "eval_runtime": 182.6215,
      "eval_samples_per_second": 0.778,
      "eval_steps_per_second": 0.778,
      "step": 3100
    },
    {
      "epoch": 6.4,
      "grad_norm": 0.9698778986930847,
      "learning_rate": 1.836734693877551e-05,
      "loss": 0.435,
      "step": 3200
    },
    {
      "epoch": 6.4,
      "eval_loss": 0.6926498413085938,
      "eval_runtime": 182.1258,
      "eval_samples_per_second": 0.78,
      "eval_steps_per_second": 0.78,
      "step": 3200
    },
    {
      "epoch": 6.6,
      "grad_norm": 1.2019530534744263,
      "learning_rate": 1.7346938775510206e-05,
      "loss": 0.4521,
      "step": 3300
    },
    {
      "epoch": 6.6,
      "eval_loss": 0.6842972636222839,
      "eval_runtime": 182.0796,
      "eval_samples_per_second": 0.78,
      "eval_steps_per_second": 0.78,
      "step": 3300
    },
    {
      "epoch": 6.8,
      "grad_norm": 1.1843819618225098,
      "learning_rate": 1.6326530612244897e-05,
      "loss": 0.4559,
      "step": 3400
    },
    {
      "epoch": 6.8,
      "eval_loss": 0.6817460656166077,
      "eval_runtime": 182.1245,
      "eval_samples_per_second": 0.78,
      "eval_steps_per_second": 0.78,
      "step": 3400
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.9009637832641602,
      "learning_rate": 1.5306122448979594e-05,
      "loss": 0.447,
      "step": 3500
    },
    {
      "epoch": 7.0,
      "eval_loss": 0.6794432401657104,
      "eval_runtime": 182.0907,
      "eval_samples_per_second": 0.78,
      "eval_steps_per_second": 0.78,
      "step": 3500
    },
    {
      "epoch": 7.2,
      "grad_norm": 1.0701977014541626,
      "learning_rate": 1.4285714285714285e-05,
      "loss": 0.3809,
      "step": 3600
    },
    {
      "epoch": 7.2,
      "eval_loss": 0.6871351599693298,
      "eval_runtime": 182.0665,
      "eval_samples_per_second": 0.78,
      "eval_steps_per_second": 0.78,
      "step": 3600
    },
    {
      "epoch": 7.4,
      "grad_norm": 0.9625453352928162,
      "learning_rate": 1.3265306122448982e-05,
      "loss": 0.3836,
      "step": 3700
    },
    {
      "epoch": 7.4,
      "eval_loss": 0.6862939596176147,
      "eval_runtime": 182.0265,
      "eval_samples_per_second": 0.78,
      "eval_steps_per_second": 0.78,
      "step": 3700
    },
    {
      "epoch": 7.6,
      "grad_norm": 1.1868125200271606,
      "learning_rate": 1.2244897959183674e-05,
      "loss": 0.3889,
      "step": 3800
    },
    {
      "epoch": 7.6,
      "eval_loss": 0.6783468723297119,
      "eval_runtime": 182.5364,
      "eval_samples_per_second": 0.778,
      "eval_steps_per_second": 0.778,
      "step": 3800
    },
    {
      "epoch": 7.8,
      "grad_norm": 0.9025134444236755,
      "learning_rate": 1.1224489795918369e-05,
      "loss": 0.4134,
      "step": 3900
    },
    {
      "epoch": 7.8,
      "eval_loss": 0.6690346002578735,
      "eval_runtime": 181.8517,
      "eval_samples_per_second": 0.781,
      "eval_steps_per_second": 0.781,
      "step": 3900
    },
    {
      "epoch": 8.0,
      "grad_norm": 1.2734243869781494,
      "learning_rate": 1.0204081632653061e-05,
      "loss": 0.4018,
      "step": 4000
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.6694031953811646,
      "eval_runtime": 182.0072,
      "eval_samples_per_second": 0.78,
      "eval_steps_per_second": 0.78,
      "step": 4000
    },
    {
      "epoch": 8.2,
      "grad_norm": 0.974176824092865,
      "learning_rate": 9.183673469387756e-06,
      "loss": 0.3526,
      "step": 4100
    },
    {
      "epoch": 8.2,
      "eval_loss": 0.6763675212860107,
      "eval_runtime": 182.0229,
      "eval_samples_per_second": 0.78,
      "eval_steps_per_second": 0.78,
      "step": 4100
    },
    {
      "epoch": 8.4,
      "grad_norm": 0.9596600532531738,
      "learning_rate": 8.163265306122448e-06,
      "loss": 0.3373,
      "step": 4200
    },
    {
      "epoch": 8.4,
      "eval_loss": 0.6745020747184753,
      "eval_runtime": 181.8856,
      "eval_samples_per_second": 0.781,
      "eval_steps_per_second": 0.781,
      "step": 4200
    },
    {
      "epoch": 8.6,
      "grad_norm": 0.9683770537376404,
      "learning_rate": 7.142857142857143e-06,
      "loss": 0.3704,
      "step": 4300
    },
    {
      "epoch": 8.6,
      "eval_loss": 0.6758388876914978,
      "eval_runtime": 181.9951,
      "eval_samples_per_second": 0.78,
      "eval_steps_per_second": 0.78,
      "step": 4300
    },
    {
      "epoch": 8.8,
      "grad_norm": 1.0410691499710083,
      "learning_rate": 6.122448979591837e-06,
      "loss": 0.3482,
      "step": 4400
    },
    {
      "epoch": 8.8,
      "eval_loss": 0.6700472831726074,
      "eval_runtime": 181.8829,
      "eval_samples_per_second": 0.781,
      "eval_steps_per_second": 0.781,
      "step": 4400
    },
    {
      "epoch": 9.0,
      "grad_norm": 1.0724315643310547,
      "learning_rate": 5.102040816326531e-06,
      "loss": 0.3621,
      "step": 4500
    },
    {
      "epoch": 9.0,
      "eval_loss": 0.66424161195755,
      "eval_runtime": 182.2196,
      "eval_samples_per_second": 0.779,
      "eval_steps_per_second": 0.779,
      "step": 4500
    },
    {
      "epoch": 9.2,
      "grad_norm": 1.2040272951126099,
      "learning_rate": 4.081632653061224e-06,
      "loss": 0.3353,
      "step": 4600
    },
    {
      "epoch": 9.2,
      "eval_loss": 0.6743778586387634,
      "eval_runtime": 182.2041,
      "eval_samples_per_second": 0.779,
      "eval_steps_per_second": 0.779,
      "step": 4600
    },
    {
      "epoch": 9.4,
      "grad_norm": 1.0811737775802612,
      "learning_rate": 3.0612244897959185e-06,
      "loss": 0.3265,
      "step": 4700
    },
    {
      "epoch": 9.4,
      "eval_loss": 0.6780717372894287,
      "eval_runtime": 182.1426,
      "eval_samples_per_second": 0.78,
      "eval_steps_per_second": 0.78,
      "step": 4700
    },
    {
      "epoch": 9.6,
      "grad_norm": 0.8568335771560669,
      "learning_rate": 2.040816326530612e-06,
      "loss": 0.3247,
      "step": 4800
    },
    {
      "epoch": 9.6,
      "eval_loss": 0.6788098216056824,
      "eval_runtime": 182.2171,
      "eval_samples_per_second": 0.779,
      "eval_steps_per_second": 0.779,
      "step": 4800
    },
    {
      "epoch": 9.8,
      "grad_norm": 1.1256827116012573,
      "learning_rate": 1.020408163265306e-06,
      "loss": 0.3355,
      "step": 4900
    },
    {
      "epoch": 9.8,
      "eval_loss": 0.6737083196640015,
      "eval_runtime": 182.1568,
      "eval_samples_per_second": 0.78,
      "eval_steps_per_second": 0.78,
      "step": 4900
    },
    {
      "epoch": 10.0,
      "grad_norm": 1.1654198169708252,
      "learning_rate": 0.0,
      "loss": 0.3152,
      "step": 5000
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.6746455430984497,
      "eval_runtime": 182.1302,
      "eval_samples_per_second": 0.78,
      "eval_steps_per_second": 0.78,
      "step": 5000
    }
  ],
  "logging_steps": 100,
  "max_steps": 5000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "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": 5.670208882951987e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}