{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 5,
  "global_step": 32,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0625,
      "grad_norm": 1.8187159299850464,
      "learning_rate": 0.00019375000000000002,
      "loss": 10.0878,
      "step": 1
    },
    {
      "epoch": 0.125,
      "grad_norm": 1.9498610496520996,
      "learning_rate": 0.0001875,
      "loss": 9.9916,
      "step": 2
    },
    {
      "epoch": 0.1875,
      "grad_norm": 2.7474117279052734,
      "learning_rate": 0.00018125000000000001,
      "loss": 9.7044,
      "step": 3
    },
    {
      "epoch": 0.25,
      "grad_norm": 3.391249418258667,
      "learning_rate": 0.000175,
      "loss": 8.9738,
      "step": 4
    },
    {
      "epoch": 0.3125,
      "grad_norm": 3.6622118949890137,
      "learning_rate": 0.00016875,
      "loss": 8.2705,
      "step": 5
    },
    {
      "epoch": 0.3125,
      "eval_clap": 0.31716692447662354,
      "eval_loss": 3.0108680725097656,
      "eval_runtime": 265.8708,
      "eval_samples_per_second": 0.03,
      "eval_steps_per_second": 0.03,
      "step": 5
    },
    {
      "epoch": 0.375,
      "grad_norm": 4.368948936462402,
      "learning_rate": 0.00016250000000000002,
      "loss": 7.8107,
      "step": 6
    },
    {
      "epoch": 0.4375,
      "grad_norm": 6.55327033996582,
      "learning_rate": 0.00015625,
      "loss": 7.507,
      "step": 7
    },
    {
      "epoch": 0.5,
      "grad_norm": 5.326369762420654,
      "learning_rate": 0.00015000000000000001,
      "loss": 7.0676,
      "step": 8
    },
    {
      "epoch": 0.5625,
      "grad_norm": 3.2963500022888184,
      "learning_rate": 0.00014375,
      "loss": 7.0271,
      "step": 9
    },
    {
      "epoch": 0.625,
      "grad_norm": 2.0462214946746826,
      "learning_rate": 0.0001375,
      "loss": 6.4565,
      "step": 10
    },
    {
      "epoch": 0.625,
      "eval_clap": 0.20497097074985504,
      "eval_loss": 3.3557252883911133,
      "eval_runtime": 503.1747,
      "eval_samples_per_second": 0.016,
      "eval_steps_per_second": 0.016,
      "step": 10
    },
    {
      "epoch": 0.6875,
      "grad_norm": 2.9796876907348633,
      "learning_rate": 0.00013125000000000002,
      "loss": 7.025,
      "step": 11
    },
    {
      "epoch": 0.75,
      "grad_norm": 1.7379363775253296,
      "learning_rate": 0.000125,
      "loss": 6.4613,
      "step": 12
    },
    {
      "epoch": 0.8125,
      "grad_norm": 2.6363821029663086,
      "learning_rate": 0.00011875,
      "loss": 6.8828,
      "step": 13
    },
    {
      "epoch": 0.875,
      "grad_norm": 1.376232624053955,
      "learning_rate": 0.00011250000000000001,
      "loss": 6.3572,
      "step": 14
    },
    {
      "epoch": 0.9375,
      "grad_norm": 1.8433048725128174,
      "learning_rate": 0.00010625000000000001,
      "loss": 6.0972,
      "step": 15
    },
    {
      "epoch": 0.9375,
      "eval_clap": 0.12289813160896301,
      "eval_loss": 4.520351409912109,
      "eval_runtime": 513.434,
      "eval_samples_per_second": 0.016,
      "eval_steps_per_second": 0.016,
      "step": 15
    },
    {
      "epoch": 1.0,
      "grad_norm": 2.2514262199401855,
      "learning_rate": 0.0001,
      "loss": 6.3686,
      "step": 16
    },
    {
      "epoch": 1.0625,
      "grad_norm": 0.9838012456893921,
      "learning_rate": 9.375e-05,
      "loss": 6.2712,
      "step": 17
    },
    {
      "epoch": 1.125,
      "grad_norm": 0.983974277973175,
      "learning_rate": 8.75e-05,
      "loss": 6.1015,
      "step": 18
    },
    {
      "epoch": 1.1875,
      "grad_norm": 0.8950835466384888,
      "learning_rate": 8.125000000000001e-05,
      "loss": 6.1462,
      "step": 19
    },
    {
      "epoch": 1.25,
      "grad_norm": 1.661649465560913,
      "learning_rate": 7.500000000000001e-05,
      "loss": 6.3825,
      "step": 20
    },
    {
      "epoch": 1.25,
      "eval_clap": 0.11188629269599915,
      "eval_loss": 4.979422092437744,
      "eval_runtime": 516.4364,
      "eval_samples_per_second": 0.015,
      "eval_steps_per_second": 0.015,
      "step": 20
    },
    {
      "epoch": 1.3125,
      "grad_norm": 1.2889989614486694,
      "learning_rate": 6.875e-05,
      "loss": 6.2617,
      "step": 21
    },
    {
      "epoch": 1.375,
      "grad_norm": 0.9493317008018494,
      "learning_rate": 6.25e-05,
      "loss": 5.9844,
      "step": 22
    },
    {
      "epoch": 1.4375,
      "grad_norm": 0.9931596517562866,
      "learning_rate": 5.6250000000000005e-05,
      "loss": 6.2795,
      "step": 23
    },
    {
      "epoch": 1.5,
      "grad_norm": 1.8261293172836304,
      "learning_rate": 5e-05,
      "loss": 5.7055,
      "step": 24
    },
    {
      "epoch": 1.5625,
      "grad_norm": 1.0377460718154907,
      "learning_rate": 4.375e-05,
      "loss": 6.3178,
      "step": 25
    },
    {
      "epoch": 1.5625,
      "eval_clap": 0.09158371388912201,
      "eval_loss": 5.108123779296875,
      "eval_runtime": 516.4832,
      "eval_samples_per_second": 0.015,
      "eval_steps_per_second": 0.015,
      "step": 25
    },
    {
      "epoch": 1.625,
      "grad_norm": 1.5140869617462158,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 6.4832,
      "step": 26
    },
    {
      "epoch": 1.6875,
      "grad_norm": 0.9569382667541504,
      "learning_rate": 3.125e-05,
      "loss": 6.3163,
      "step": 27
    },
    {
      "epoch": 1.75,
      "grad_norm": 1.586517095565796,
      "learning_rate": 2.5e-05,
      "loss": 6.2888,
      "step": 28
    },
    {
      "epoch": 1.8125,
      "grad_norm": 1.238013744354248,
      "learning_rate": 1.8750000000000002e-05,
      "loss": 6.1361,
      "step": 29
    },
    {
      "epoch": 1.875,
      "grad_norm": 1.0142874717712402,
      "learning_rate": 1.25e-05,
      "loss": 6.1301,
      "step": 30
    },
    {
      "epoch": 1.875,
      "eval_clap": 0.08499659597873688,
      "eval_loss": 5.103597640991211,
      "eval_runtime": 424.9013,
      "eval_samples_per_second": 0.019,
      "eval_steps_per_second": 0.019,
      "step": 30
    },
    {
      "epoch": 1.9375,
      "grad_norm": 0.9664702415466309,
      "learning_rate": 6.25e-06,
      "loss": 6.3296,
      "step": 31
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.8057743310928345,
      "learning_rate": 0.0,
      "loss": 6.544,
      "step": 32
    },
    {
      "epoch": 2.0,
      "step": 32,
      "total_flos": 152209974742572.0,
      "train_loss": 6.930231094360352,
      "train_runtime": 3130.5046,
      "train_samples_per_second": 0.077,
      "train_steps_per_second": 0.01
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 32,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 152209974742572.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}