{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.9981185324553152,
  "eval_steps": 500,
  "global_step": 531,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04,
      "grad_norm": 2.453125,
      "learning_rate": 2.5e-06,
      "loss": 1.6646,
      "step": 10
    },
    {
      "epoch": 0.08,
      "grad_norm": 2.71875,
      "learning_rate": 5e-06,
      "loss": 1.633,
      "step": 20
    },
    {
      "epoch": 0.11,
      "grad_norm": 1.453125,
      "learning_rate": 7.500000000000001e-06,
      "loss": 1.5999,
      "step": 30
    },
    {
      "epoch": 0.15,
      "grad_norm": 1.921875,
      "learning_rate": 1e-05,
      "loss": 1.6187,
      "step": 40
    },
    {
      "epoch": 0.19,
      "grad_norm": 2.625,
      "learning_rate": 1.25e-05,
      "loss": 1.5918,
      "step": 50
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.59765625,
      "learning_rate": 1.5000000000000002e-05,
      "loss": 1.5936,
      "step": 60
    },
    {
      "epoch": 0.26,
      "grad_norm": 2.015625,
      "learning_rate": 1.7500000000000002e-05,
      "loss": 1.5263,
      "step": 70
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.92578125,
      "learning_rate": 2e-05,
      "loss": 1.4784,
      "step": 80
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.474609375,
      "learning_rate": 1.999034865600726e-05,
      "loss": 1.3839,
      "step": 90
    },
    {
      "epoch": 0.38,
      "grad_norm": 1.2109375,
      "learning_rate": 1.9961413253717214e-05,
      "loss": 1.3698,
      "step": 100
    },
    {
      "epoch": 0.41,
      "grad_norm": 1.0390625,
      "learning_rate": 1.9913249646234072e-05,
      "loss": 1.2916,
      "step": 110
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.35546875,
      "learning_rate": 1.9845950802266584e-05,
      "loss": 1.2607,
      "step": 120
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.375,
      "learning_rate": 1.9759646626673445e-05,
      "loss": 1.2405,
      "step": 130
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.8828125,
      "learning_rate": 1.9654503709711984e-05,
      "loss": 1.2615,
      "step": 140
    },
    {
      "epoch": 0.56,
      "grad_norm": 1.2421875,
      "learning_rate": 1.9530725005474195e-05,
      "loss": 1.1433,
      "step": 150
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.404296875,
      "learning_rate": 1.93885494401308e-05,
      "loss": 1.1599,
      "step": 160
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.91796875,
      "learning_rate": 1.9228251450739495e-05,
      "loss": 1.1325,
      "step": 170
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.55859375,
      "learning_rate": 1.905014045550767e-05,
      "loss": 1.0797,
      "step": 180
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.375,
      "learning_rate": 1.8854560256532098e-05,
      "loss": 1.041,
      "step": 190
    },
    {
      "epoch": 0.75,
      "grad_norm": 1.2265625,
      "learning_rate": 1.8641888376168483e-05,
      "loss": 1.0286,
      "step": 200
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.287109375,
      "learning_rate": 1.8412535328311813e-05,
      "loss": 0.9831,
      "step": 210
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.28515625,
      "learning_rate": 1.816694382599422e-05,
      "loss": 0.9927,
      "step": 220
    },
    {
      "epoch": 0.87,
      "grad_norm": 3.25,
      "learning_rate": 1.7905587926829815e-05,
      "loss": 1.0075,
      "step": 230
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.255859375,
      "learning_rate": 1.762897211795607e-05,
      "loss": 1.0232,
      "step": 240
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.6484375,
      "learning_rate": 1.733763034223804e-05,
      "loss": 0.9476,
      "step": 250
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.67578125,
      "learning_rate": 1.7032124967615112e-05,
      "loss": 0.9338,
      "step": 260
    },
    {
      "epoch": 1.02,
      "grad_norm": 0.41796875,
      "learning_rate": 1.6713045701579705e-05,
      "loss": 0.9243,
      "step": 270
    },
    {
      "epoch": 1.05,
      "grad_norm": 0.232421875,
      "learning_rate": 1.638100845288331e-05,
      "loss": 0.9084,
      "step": 280
    },
    {
      "epoch": 1.09,
      "grad_norm": 0.349609375,
      "learning_rate": 1.6036654142667043e-05,
      "loss": 0.896,
      "step": 290
    },
    {
      "epoch": 1.13,
      "grad_norm": 0.25390625,
      "learning_rate": 1.568064746731156e-05,
      "loss": 0.9311,
      "step": 300
    },
    {
      "epoch": 1.17,
      "grad_norm": 0.236328125,
      "learning_rate": 1.5313675615394373e-05,
      "loss": 0.9182,
      "step": 310
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.326171875,
      "learning_rate": 1.4936446941231186e-05,
      "loss": 0.8967,
      "step": 320
    },
    {
      "epoch": 1.24,
      "grad_norm": 0.2451171875,
      "learning_rate": 1.4549689597561652e-05,
      "loss": 0.8965,
      "step": 330
    },
    {
      "epoch": 1.28,
      "grad_norm": 0.435546875,
      "learning_rate": 1.4154150130018867e-05,
      "loss": 0.9105,
      "step": 340
    },
    {
      "epoch": 1.32,
      "grad_norm": 0.3359375,
      "learning_rate": 1.375059203609562e-05,
      "loss": 0.9089,
      "step": 350
    },
    {
      "epoch": 1.35,
      "grad_norm": 0.2578125,
      "learning_rate": 1.3339794291389015e-05,
      "loss": 0.9066,
      "step": 360
    },
    {
      "epoch": 1.39,
      "grad_norm": 0.271484375,
      "learning_rate": 1.2922549845968174e-05,
      "loss": 0.8878,
      "step": 370
    },
    {
      "epoch": 1.43,
      "grad_norm": 0.30078125,
      "learning_rate": 1.2499664093767458e-05,
      "loss": 0.8909,
      "step": 380
    },
    {
      "epoch": 1.47,
      "grad_norm": 2.8125,
      "learning_rate": 1.2071953317959692e-05,
      "loss": 0.8659,
      "step": 390
    },
    {
      "epoch": 1.51,
      "grad_norm": 0.3125,
      "learning_rate": 1.1640243115310219e-05,
      "loss": 0.8685,
      "step": 400
    },
    {
      "epoch": 1.54,
      "grad_norm": 0.216796875,
      "learning_rate": 1.1205366802553231e-05,
      "loss": 0.8959,
      "step": 410
    },
    {
      "epoch": 1.58,
      "grad_norm": 0.310546875,
      "learning_rate": 1.076816380786647e-05,
      "loss": 0.9412,
      "step": 420
    },
    {
      "epoch": 1.62,
      "grad_norm": 0.99609375,
      "learning_rate": 1.0329478050549208e-05,
      "loss": 0.8929,
      "step": 430
    },
    {
      "epoch": 1.66,
      "grad_norm": 0.3359375,
      "learning_rate": 9.890156312031165e-06,
      "loss": 0.9102,
      "step": 440
    },
    {
      "epoch": 1.69,
      "grad_norm": 0.37109375,
      "learning_rate": 9.451046601356725e-06,
      "loss": 0.8645,
      "step": 450
    },
    {
      "epoch": 1.73,
      "grad_norm": 0.259765625,
      "learning_rate": 9.012996518299547e-06,
      "loss": 0.8554,
      "step": 460
    },
    {
      "epoch": 1.77,
      "grad_norm": 0.314453125,
      "learning_rate": 8.576851617267151e-06,
      "loss": 0.9034,
      "step": 470
    },
    {
      "epoch": 1.81,
      "grad_norm": 0.2001953125,
      "learning_rate": 8.143453775153646e-06,
      "loss": 0.8696,
      "step": 480
    },
    {
      "epoch": 1.84,
      "grad_norm": 0.671875,
      "learning_rate": 7.713639566291028e-06,
      "loss": 0.8801,
      "step": 490
    },
    {
      "epoch": 1.88,
      "grad_norm": 0.291015625,
      "learning_rate": 7.2882386476358304e-06,
      "loss": 0.8872,
      "step": 500
    },
    {
      "epoch": 1.92,
      "grad_norm": 0.37109375,
      "learning_rate": 6.868072157308213e-06,
      "loss": 0.8586,
      "step": 510
    },
    {
      "epoch": 1.96,
      "grad_norm": 0.7109375,
      "learning_rate": 6.453951129574644e-06,
      "loss": 0.8727,
      "step": 520
    },
    {
      "epoch": 1.99,
      "grad_norm": 1.15625,
      "learning_rate": 6.046674929333787e-06,
      "loss": 0.8833,
      "step": 530
    }
  ],
  "logging_steps": 10,
  "max_steps": 795,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 1.03886006255616e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}