{
  "best_metric": 0.8340622782707214,
  "best_model_checkpoint": "./results/checkpoint-1000",
  "epoch": 2.9940436796823295,
  "eval_steps": 500,
  "global_step": 1131,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05294506949040371,
      "grad_norm": 62.576904296875,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 16.7099,
      "step": 20
    },
    {
      "epoch": 0.10589013898080742,
      "grad_norm": 35.65088653564453,
      "learning_rate": 4.000000000000001e-06,
      "loss": 14.7891,
      "step": 40
    },
    {
      "epoch": 0.15883520847121113,
      "grad_norm": 30.480260848999023,
      "learning_rate": 6e-06,
      "loss": 11.6833,
      "step": 60
    },
    {
      "epoch": 0.21178027796161483,
      "grad_norm": 24.958763122558594,
      "learning_rate": 8.000000000000001e-06,
      "loss": 7.878,
      "step": 80
    },
    {
      "epoch": 0.26472534745201853,
      "grad_norm": 4.896631717681885,
      "learning_rate": 1e-05,
      "loss": 4.1299,
      "step": 100
    },
    {
      "epoch": 0.31767041694242226,
      "grad_norm": 1.8821851015090942,
      "learning_rate": 1.2e-05,
      "loss": 2.512,
      "step": 120
    },
    {
      "epoch": 0.37061548643282594,
      "grad_norm": 0.7400406002998352,
      "learning_rate": 1.4000000000000001e-05,
      "loss": 1.8807,
      "step": 140
    },
    {
      "epoch": 0.42356055592322966,
      "grad_norm": 0.601634681224823,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 1.5617,
      "step": 160
    },
    {
      "epoch": 0.47650562541363334,
      "grad_norm": 0.5250083208084106,
      "learning_rate": 1.8e-05,
      "loss": 1.461,
      "step": 180
    },
    {
      "epoch": 0.5294506949040371,
      "grad_norm": 0.5037821531295776,
      "learning_rate": 2e-05,
      "loss": 1.3703,
      "step": 200
    },
    {
      "epoch": 0.5823957643944407,
      "grad_norm": 0.47873854637145996,
      "learning_rate": 2.2000000000000003e-05,
      "loss": 1.2966,
      "step": 220
    },
    {
      "epoch": 0.6353408338848445,
      "grad_norm": 3.9664485454559326,
      "learning_rate": 2.4e-05,
      "loss": 1.2427,
      "step": 240
    },
    {
      "epoch": 0.6882859033752482,
      "grad_norm": 0.45016345381736755,
      "learning_rate": 2.6000000000000002e-05,
      "loss": 1.2139,
      "step": 260
    },
    {
      "epoch": 0.7412309728656519,
      "grad_norm": 0.5136398077011108,
      "learning_rate": 2.8000000000000003e-05,
      "loss": 1.1679,
      "step": 280
    },
    {
      "epoch": 0.7941760423560555,
      "grad_norm": 0.3736862242221832,
      "learning_rate": 3e-05,
      "loss": 1.1604,
      "step": 300
    },
    {
      "epoch": 0.8471211118464593,
      "grad_norm": 0.38698282837867737,
      "learning_rate": 3.2000000000000005e-05,
      "loss": 1.1407,
      "step": 320
    },
    {
      "epoch": 0.900066181336863,
      "grad_norm": 0.4257580637931824,
      "learning_rate": 3.4000000000000007e-05,
      "loss": 1.1177,
      "step": 340
    },
    {
      "epoch": 0.9530112508272667,
      "grad_norm": 0.3982521891593933,
      "learning_rate": 3.6e-05,
      "loss": 1.1199,
      "step": 360
    },
    {
      "epoch": 1.0059563203176705,
      "grad_norm": 0.3849237859249115,
      "learning_rate": 3.8e-05,
      "loss": 1.0925,
      "step": 380
    },
    {
      "epoch": 1.0589013898080741,
      "grad_norm": 0.3753887414932251,
      "learning_rate": 4e-05,
      "loss": 1.0605,
      "step": 400
    },
    {
      "epoch": 1.1118464592984778,
      "grad_norm": 0.3810591697692871,
      "learning_rate": 4.2e-05,
      "loss": 1.0541,
      "step": 420
    },
    {
      "epoch": 1.1647915287888815,
      "grad_norm": 0.3707886040210724,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 1.0334,
      "step": 440
    },
    {
      "epoch": 1.2177365982792852,
      "grad_norm": 0.36902502179145813,
      "learning_rate": 4.600000000000001e-05,
      "loss": 1.0249,
      "step": 460
    },
    {
      "epoch": 1.270681667769689,
      "grad_norm": 0.3862062692642212,
      "learning_rate": 4.8e-05,
      "loss": 1.0141,
      "step": 480
    },
    {
      "epoch": 1.3236267372600927,
      "grad_norm": 0.36468520760536194,
      "learning_rate": 5e-05,
      "loss": 1.063,
      "step": 500
    },
    {
      "epoch": 1.3236267372600927,
      "eval_loss": 0.9045532941818237,
      "eval_runtime": 74.2771,
      "eval_samples_per_second": 10.165,
      "eval_steps_per_second": 2.545,
      "step": 500
    },
    {
      "epoch": 1.3765718067504964,
      "grad_norm": 0.3426459729671478,
      "learning_rate": 4.8415213946117275e-05,
      "loss": 1.01,
      "step": 520
    },
    {
      "epoch": 1.4295168762409,
      "grad_norm": 0.3736313581466675,
      "learning_rate": 4.6830427892234554e-05,
      "loss": 0.984,
      "step": 540
    },
    {
      "epoch": 1.4824619457313037,
      "grad_norm": 0.36285898089408875,
      "learning_rate": 4.524564183835183e-05,
      "loss": 0.9816,
      "step": 560
    },
    {
      "epoch": 1.5354070152217076,
      "grad_norm": 0.37838441133499146,
      "learning_rate": 4.36608557844691e-05,
      "loss": 0.9807,
      "step": 580
    },
    {
      "epoch": 1.588352084712111,
      "grad_norm": 0.3449678421020508,
      "learning_rate": 4.207606973058637e-05,
      "loss": 0.982,
      "step": 600
    },
    {
      "epoch": 1.641297154202515,
      "grad_norm": 0.3467804789543152,
      "learning_rate": 4.0491283676703644e-05,
      "loss": 0.9553,
      "step": 620
    },
    {
      "epoch": 1.6942422236929184,
      "grad_norm": 0.3551880419254303,
      "learning_rate": 3.8906497622820917e-05,
      "loss": 0.9701,
      "step": 640
    },
    {
      "epoch": 1.7471872931833223,
      "grad_norm": 0.3425547182559967,
      "learning_rate": 3.7321711568938196e-05,
      "loss": 0.973,
      "step": 660
    },
    {
      "epoch": 1.800132362673726,
      "grad_norm": 0.32189810276031494,
      "learning_rate": 3.573692551505547e-05,
      "loss": 0.9599,
      "step": 680
    },
    {
      "epoch": 1.8530774321641297,
      "grad_norm": 0.34214696288108826,
      "learning_rate": 3.415213946117274e-05,
      "loss": 0.952,
      "step": 700
    },
    {
      "epoch": 1.9060225016545336,
      "grad_norm": 0.33412784337997437,
      "learning_rate": 3.256735340729002e-05,
      "loss": 0.9453,
      "step": 720
    },
    {
      "epoch": 1.958967571144937,
      "grad_norm": 0.33273905515670776,
      "learning_rate": 3.098256735340729e-05,
      "loss": 0.9387,
      "step": 740
    },
    {
      "epoch": 2.011912640635341,
      "grad_norm": 0.32698702812194824,
      "learning_rate": 2.939778129952457e-05,
      "loss": 0.9456,
      "step": 760
    },
    {
      "epoch": 2.0648577101257444,
      "grad_norm": 0.36428529024124146,
      "learning_rate": 2.7812995245641837e-05,
      "loss": 0.9445,
      "step": 780
    },
    {
      "epoch": 2.1178027796161483,
      "grad_norm": 0.3233266770839691,
      "learning_rate": 2.6228209191759113e-05,
      "loss": 0.9185,
      "step": 800
    },
    {
      "epoch": 2.1707478491065517,
      "grad_norm": 0.3173067569732666,
      "learning_rate": 2.4643423137876386e-05,
      "loss": 0.9146,
      "step": 820
    },
    {
      "epoch": 2.2236929185969556,
      "grad_norm": 0.33917301893234253,
      "learning_rate": 2.305863708399366e-05,
      "loss": 0.9195,
      "step": 840
    },
    {
      "epoch": 2.2766379880873595,
      "grad_norm": 0.3438282608985901,
      "learning_rate": 2.1473851030110938e-05,
      "loss": 0.9356,
      "step": 860
    },
    {
      "epoch": 2.329583057577763,
      "grad_norm": 0.33590319752693176,
      "learning_rate": 1.988906497622821e-05,
      "loss": 0.9319,
      "step": 880
    },
    {
      "epoch": 2.382528127068167,
      "grad_norm": 0.5414553880691528,
      "learning_rate": 1.8304278922345483e-05,
      "loss": 0.9208,
      "step": 900
    },
    {
      "epoch": 2.4354731965585703,
      "grad_norm": 0.34509792923927307,
      "learning_rate": 1.671949286846276e-05,
      "loss": 0.9026,
      "step": 920
    },
    {
      "epoch": 2.488418266048974,
      "grad_norm": 0.30984020233154297,
      "learning_rate": 1.5134706814580033e-05,
      "loss": 0.9066,
      "step": 940
    },
    {
      "epoch": 2.541363335539378,
      "grad_norm": 0.31895536184310913,
      "learning_rate": 1.3549920760697307e-05,
      "loss": 0.9275,
      "step": 960
    },
    {
      "epoch": 2.5943084050297816,
      "grad_norm": 0.3005692660808563,
      "learning_rate": 1.1965134706814581e-05,
      "loss": 0.9241,
      "step": 980
    },
    {
      "epoch": 2.6472534745201854,
      "grad_norm": 0.325959712266922,
      "learning_rate": 1.0380348652931855e-05,
      "loss": 0.9181,
      "step": 1000
    },
    {
      "epoch": 2.6472534745201854,
      "eval_loss": 0.8340622782707214,
      "eval_runtime": 74.3137,
      "eval_samples_per_second": 10.16,
      "eval_steps_per_second": 2.543,
      "step": 1000
    },
    {
      "epoch": 2.700198544010589,
      "grad_norm": 0.37024152278900146,
      "learning_rate": 8.79556259904913e-06,
      "loss": 0.9156,
      "step": 1020
    },
    {
      "epoch": 2.753143613500993,
      "grad_norm": 0.3021298050880432,
      "learning_rate": 7.2107765451664034e-06,
      "loss": 0.9204,
      "step": 1040
    },
    {
      "epoch": 2.8060886829913967,
      "grad_norm": 0.35478419065475464,
      "learning_rate": 5.625990491283677e-06,
      "loss": 0.9342,
      "step": 1060
    },
    {
      "epoch": 2.8590337524818,
      "grad_norm": 0.34113916754722595,
      "learning_rate": 4.041204437400952e-06,
      "loss": 0.9187,
      "step": 1080
    },
    {
      "epoch": 2.9119788219722036,
      "grad_norm": 0.33516696095466614,
      "learning_rate": 2.456418383518225e-06,
      "loss": 0.9238,
      "step": 1100
    },
    {
      "epoch": 2.9649238914626075,
      "grad_norm": 0.2989369034767151,
      "learning_rate": 8.716323296354993e-07,
      "loss": 0.9188,
      "step": 1120
    }
  ],
  "logging_steps": 20,
  "max_steps": 1131,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.101970198757376e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}