{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.020380434782608696,
  "eval_steps": 500,
  "global_step": 75,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0005434782608695652,
      "grad_norm": 0.03744836896657944,
      "learning_rate": 4.966666666666667e-06,
      "loss": 0.0091,
      "on_policy_loss": 0.0091,
      "step": 2
    },
    {
      "epoch": 0.0010869565217391304,
      "grad_norm": 0.029108790680766106,
      "learning_rate": 4.9000000000000005e-06,
      "loss": 0.0097,
      "on_policy_loss": 0.0097,
      "step": 4
    },
    {
      "epoch": 0.0016304347826086956,
      "grad_norm": 0.02633393555879593,
      "learning_rate": 4.833333333333333e-06,
      "loss": 0.013,
      "on_policy_loss": 0.013,
      "step": 6
    },
    {
      "epoch": 0.002173913043478261,
      "grad_norm": 0.018148666247725487,
      "learning_rate": 4.766666666666667e-06,
      "loss": 0.0099,
      "on_policy_loss": 0.0099,
      "step": 8
    },
    {
      "epoch": 0.002717391304347826,
      "grad_norm": 0.02139461226761341,
      "learning_rate": 4.7e-06,
      "loss": 0.0118,
      "on_policy_loss": 0.0118,
      "step": 10
    },
    {
      "epoch": 0.003260869565217391,
      "grad_norm": 0.028195466846227646,
      "learning_rate": 4.633333333333334e-06,
      "loss": 0.0112,
      "on_policy_loss": 0.0112,
      "step": 12
    },
    {
      "epoch": 0.0038043478260869567,
      "grad_norm": 0.0212844368070364,
      "learning_rate": 4.566666666666667e-06,
      "loss": 0.0114,
      "on_policy_loss": 0.0114,
      "step": 14
    },
    {
      "epoch": 0.004347826086956522,
      "grad_norm": 0.03395017609000206,
      "learning_rate": 4.5e-06,
      "loss": 0.0109,
      "on_policy_loss": 0.0109,
      "step": 16
    },
    {
      "epoch": 0.004891304347826087,
      "grad_norm": 0.03911148011684418,
      "learning_rate": 4.433333333333334e-06,
      "loss": 0.009,
      "on_policy_loss": 0.009,
      "step": 18
    },
    {
      "epoch": 0.005434782608695652,
      "grad_norm": 0.01619447022676468,
      "learning_rate": 4.366666666666667e-06,
      "loss": 0.0133,
      "on_policy_loss": 0.0133,
      "step": 20
    },
    {
      "epoch": 0.005978260869565218,
      "grad_norm": 0.02550458535552025,
      "learning_rate": 4.3e-06,
      "loss": 0.0104,
      "on_policy_loss": 0.0104,
      "step": 22
    },
    {
      "epoch": 0.006521739130434782,
      "grad_norm": 0.03460410609841347,
      "learning_rate": 4.233333333333334e-06,
      "loss": 0.0099,
      "on_policy_loss": 0.0099,
      "step": 24
    },
    {
      "epoch": 0.007065217391304348,
      "grad_norm": 0.031486913561820984,
      "learning_rate": 4.166666666666667e-06,
      "loss": 0.0128,
      "on_policy_loss": 0.0128,
      "step": 26
    },
    {
      "epoch": 0.007608695652173913,
      "grad_norm": 0.022056138142943382,
      "learning_rate": 4.1e-06,
      "loss": 0.0063,
      "on_policy_loss": 0.0063,
      "step": 28
    },
    {
      "epoch": 0.008152173913043478,
      "grad_norm": 0.016603024676442146,
      "learning_rate": 4.033333333333333e-06,
      "loss": 0.0074,
      "on_policy_loss": 0.0074,
      "step": 30
    },
    {
      "epoch": 0.008695652173913044,
      "grad_norm": 0.017155027016997337,
      "learning_rate": 3.966666666666667e-06,
      "loss": 0.0125,
      "on_policy_loss": 0.0125,
      "step": 32
    },
    {
      "epoch": 0.00923913043478261,
      "grad_norm": 0.022139370441436768,
      "learning_rate": 3.900000000000001e-06,
      "loss": 0.0091,
      "on_policy_loss": 0.0091,
      "step": 34
    },
    {
      "epoch": 0.009782608695652175,
      "grad_norm": 0.019025282934308052,
      "learning_rate": 3.833333333333334e-06,
      "loss": 0.0088,
      "on_policy_loss": 0.0088,
      "step": 36
    },
    {
      "epoch": 0.010326086956521738,
      "grad_norm": 0.018911313265562057,
      "learning_rate": 3.766666666666667e-06,
      "loss": 0.0088,
      "on_policy_loss": 0.0088,
      "step": 38
    },
    {
      "epoch": 0.010869565217391304,
      "grad_norm": 0.021639954298734665,
      "learning_rate": 3.7e-06,
      "loss": 0.0058,
      "on_policy_loss": 0.0058,
      "step": 40
    },
    {
      "epoch": 0.01141304347826087,
      "grad_norm": 0.01750766858458519,
      "learning_rate": 3.633333333333334e-06,
      "loss": 0.0101,
      "on_policy_loss": 0.0101,
      "step": 42
    },
    {
      "epoch": 0.011956521739130435,
      "grad_norm": 0.018655167892575264,
      "learning_rate": 3.566666666666667e-06,
      "loss": 0.0057,
      "on_policy_loss": 0.0057,
      "step": 44
    },
    {
      "epoch": 0.0125,
      "grad_norm": 0.017098240554332733,
      "learning_rate": 3.5e-06,
      "loss": 0.0051,
      "on_policy_loss": 0.0051,
      "step": 46
    },
    {
      "epoch": 0.013043478260869565,
      "grad_norm": 0.01902070641517639,
      "learning_rate": 3.4333333333333336e-06,
      "loss": 0.0066,
      "on_policy_loss": 0.0066,
      "step": 48
    },
    {
      "epoch": 0.01358695652173913,
      "grad_norm": 0.018735554069280624,
      "learning_rate": 3.366666666666667e-06,
      "loss": 0.0102,
      "on_policy_loss": 0.0102,
      "step": 50
    },
    {
      "epoch": 0.014130434782608696,
      "grad_norm": 0.01960255578160286,
      "learning_rate": 3.3000000000000006e-06,
      "loss": 0.0091,
      "on_policy_loss": 0.0091,
      "step": 52
    },
    {
      "epoch": 0.014673913043478261,
      "grad_norm": 0.017713133245706558,
      "learning_rate": 3.2333333333333334e-06,
      "loss": 0.0099,
      "on_policy_loss": 0.0099,
      "step": 54
    },
    {
      "epoch": 0.015217391304347827,
      "grad_norm": 0.017586251720786095,
      "learning_rate": 3.1666666666666667e-06,
      "loss": 0.0068,
      "on_policy_loss": 0.0068,
      "step": 56
    },
    {
      "epoch": 0.01576086956521739,
      "grad_norm": 0.018474629148840904,
      "learning_rate": 3.1000000000000004e-06,
      "loss": 0.0114,
      "on_policy_loss": 0.0114,
      "step": 58
    },
    {
      "epoch": 0.016304347826086956,
      "grad_norm": 0.01840766705572605,
      "learning_rate": 3.0333333333333337e-06,
      "loss": 0.0056,
      "on_policy_loss": 0.0056,
      "step": 60
    },
    {
      "epoch": 0.01684782608695652,
      "grad_norm": 0.017008541151881218,
      "learning_rate": 2.9666666666666673e-06,
      "loss": 0.0068,
      "on_policy_loss": 0.0068,
      "step": 62
    },
    {
      "epoch": 0.017391304347826087,
      "grad_norm": 0.017770633101463318,
      "learning_rate": 2.9e-06,
      "loss": 0.0083,
      "on_policy_loss": 0.0083,
      "step": 64
    },
    {
      "epoch": 0.017934782608695653,
      "grad_norm": 0.017109453678131104,
      "learning_rate": 2.8333333333333335e-06,
      "loss": 0.0071,
      "on_policy_loss": 0.0071,
      "step": 66
    },
    {
      "epoch": 0.01847826086956522,
      "grad_norm": 0.01635884866118431,
      "learning_rate": 2.766666666666667e-06,
      "loss": 0.0087,
      "on_policy_loss": 0.0087,
      "step": 68
    },
    {
      "epoch": 0.019021739130434784,
      "grad_norm": 0.01571054942905903,
      "learning_rate": 2.7000000000000004e-06,
      "loss": 0.0085,
      "on_policy_loss": 0.0085,
      "step": 70
    },
    {
      "epoch": 0.01956521739130435,
      "grad_norm": 0.01731751672923565,
      "learning_rate": 2.6333333333333332e-06,
      "loss": 0.0065,
      "on_policy_loss": 0.0065,
      "step": 72
    },
    {
      "epoch": 0.02010869565217391,
      "grad_norm": 0.015707146376371384,
      "learning_rate": 2.566666666666667e-06,
      "loss": 0.0085,
      "on_policy_loss": 0.0085,
      "step": 74
    }
  ],
  "logging_steps": 2,
  "max_steps": 150,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 25,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}