{
  "best_metric": 1.2785857915878296,
  "best_model_checkpoint": "outputs/checkpoint-460",
  "epoch": 9.932523616734143,
  "eval_steps": 500,
  "global_step": 460,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.4318488529014845,
      "grad_norm": 1.2942979335784912,
      "learning_rate": 6e-06,
      "loss": 2.9144,
      "step": 20
    },
    {
      "epoch": 0.863697705802969,
      "grad_norm": 1.3759782314300537,
      "learning_rate": 1.2e-05,
      "loss": 2.8035,
      "step": 40
    },
    {
      "epoch": 0.9932523616734144,
      "eval_loss": 2.4753777980804443,
      "eval_runtime": 12.0523,
      "eval_samples_per_second": 30.865,
      "eval_steps_per_second": 3.9,
      "step": 46
    },
    {
      "epoch": 1.2955465587044535,
      "grad_norm": 1.218517541885376,
      "learning_rate": 1.8e-05,
      "loss": 2.5059,
      "step": 60
    },
    {
      "epoch": 1.7273954116059378,
      "grad_norm": 0.5991122722625732,
      "learning_rate": 2.4e-05,
      "loss": 2.0867,
      "step": 80
    },
    {
      "epoch": 1.9865047233468287,
      "eval_loss": 1.8320975303649902,
      "eval_runtime": 12.0828,
      "eval_samples_per_second": 30.787,
      "eval_steps_per_second": 3.89,
      "step": 92
    },
    {
      "epoch": 2.1592442645074224,
      "grad_norm": 0.5117250680923462,
      "learning_rate": 3e-05,
      "loss": 2.0309,
      "step": 100
    },
    {
      "epoch": 2.591093117408907,
      "grad_norm": 0.4918386936187744,
      "learning_rate": 2.977211629518312e-05,
      "loss": 1.9261,
      "step": 120
    },
    {
      "epoch": 2.979757085020243,
      "eval_loss": 1.6659022569656372,
      "eval_runtime": 12.0832,
      "eval_samples_per_second": 30.787,
      "eval_steps_per_second": 3.89,
      "step": 138
    },
    {
      "epoch": 3.0229419703103915,
      "grad_norm": 0.7239399552345276,
      "learning_rate": 2.9095389311788626e-05,
      "loss": 1.917,
      "step": 140
    },
    {
      "epoch": 3.454790823211876,
      "grad_norm": 0.745520830154419,
      "learning_rate": 2.7990381056766583e-05,
      "loss": 1.8193,
      "step": 160
    },
    {
      "epoch": 3.8866396761133606,
      "grad_norm": 0.7848306894302368,
      "learning_rate": 2.649066664678467e-05,
      "loss": 1.7836,
      "step": 180
    },
    {
      "epoch": 3.9946018893387314,
      "eval_loss": 1.5325894355773926,
      "eval_runtime": 12.0426,
      "eval_samples_per_second": 30.89,
      "eval_steps_per_second": 3.903,
      "step": 185
    },
    {
      "epoch": 4.318488529014845,
      "grad_norm": 1.0365893840789795,
      "learning_rate": 2.464181414529809e-05,
      "loss": 1.708,
      "step": 200
    },
    {
      "epoch": 4.75033738191633,
      "grad_norm": 1.2468961477279663,
      "learning_rate": 2.25e-05,
      "loss": 1.7037,
      "step": 220
    },
    {
      "epoch": 4.987854251012146,
      "eval_loss": 1.4344478845596313,
      "eval_runtime": 12.0914,
      "eval_samples_per_second": 30.766,
      "eval_steps_per_second": 3.887,
      "step": 231
    },
    {
      "epoch": 5.182186234817814,
      "grad_norm": 1.0612260103225708,
      "learning_rate": 2.0130302149885033e-05,
      "loss": 1.6629,
      "step": 240
    },
    {
      "epoch": 5.614035087719298,
      "grad_norm": 1.2603241205215454,
      "learning_rate": 1.760472266500396e-05,
      "loss": 1.6026,
      "step": 260
    },
    {
      "epoch": 5.98110661268556,
      "eval_loss": 1.3610283136367798,
      "eval_runtime": 12.085,
      "eval_samples_per_second": 30.782,
      "eval_steps_per_second": 3.889,
      "step": 277
    },
    {
      "epoch": 6.045883940620783,
      "grad_norm": 1.312119722366333,
      "learning_rate": 1.5e-05,
      "loss": 1.5725,
      "step": 280
    },
    {
      "epoch": 6.477732793522267,
      "grad_norm": 1.3641616106033325,
      "learning_rate": 1.2395277334996045e-05,
      "loss": 1.5075,
      "step": 300
    },
    {
      "epoch": 6.909581646423752,
      "grad_norm": 1.4151686429977417,
      "learning_rate": 9.86969785011497e-06,
      "loss": 1.5518,
      "step": 320
    },
    {
      "epoch": 6.995951417004049,
      "eval_loss": 1.3124128580093384,
      "eval_runtime": 12.0879,
      "eval_samples_per_second": 30.775,
      "eval_steps_per_second": 3.888,
      "step": 324
    },
    {
      "epoch": 7.341430499325236,
      "grad_norm": 1.6003135442733765,
      "learning_rate": 7.500000000000004e-06,
      "loss": 1.4876,
      "step": 340
    },
    {
      "epoch": 7.77327935222672,
      "grad_norm": 1.3969717025756836,
      "learning_rate": 5.3581858547019095e-06,
      "loss": 1.5052,
      "step": 360
    },
    {
      "epoch": 7.989203778677463,
      "eval_loss": 1.288710594177246,
      "eval_runtime": 12.0435,
      "eval_samples_per_second": 30.888,
      "eval_steps_per_second": 3.903,
      "step": 370
    },
    {
      "epoch": 8.205128205128204,
      "grad_norm": 1.4700652360916138,
      "learning_rate": 3.5093333532153316e-06,
      "loss": 1.48,
      "step": 380
    },
    {
      "epoch": 8.63697705802969,
      "grad_norm": 1.7552611827850342,
      "learning_rate": 2.0096189432334194e-06,
      "loss": 1.4669,
      "step": 400
    },
    {
      "epoch": 8.982456140350877,
      "eval_loss": 1.2798781394958496,
      "eval_runtime": 12.0997,
      "eval_samples_per_second": 30.744,
      "eval_steps_per_second": 3.884,
      "step": 416
    },
    {
      "epoch": 9.068825910931174,
      "grad_norm": 1.3221427202224731,
      "learning_rate": 9.046106882113753e-07,
      "loss": 1.4745,
      "step": 420
    },
    {
      "epoch": 9.50067476383266,
      "grad_norm": 1.6423956155776978,
      "learning_rate": 2.278837048168797e-07,
      "loss": 1.4449,
      "step": 440
    },
    {
      "epoch": 9.932523616734143,
      "grad_norm": 1.3725236654281616,
      "learning_rate": 0.0,
      "loss": 1.4768,
      "step": 460
    },
    {
      "epoch": 9.932523616734143,
      "eval_loss": 1.2785857915878296,
      "eval_runtime": 12.0719,
      "eval_samples_per_second": 30.815,
      "eval_steps_per_second": 3.893,
      "step": 460
    }
  ],
  "logging_steps": 20,
  "max_steps": 460,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 4.088498313476506e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}