{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.111111111111111,
  "eval_steps": 2,
  "global_step": 14,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2222222222222222,
      "grad_norm": 1.2958176136016846,
      "learning_rate": 1e-05,
      "loss": 1.3685,
      "step": 1
    },
    {
      "epoch": 0.2222222222222222,
      "eval_loss": 1.563645601272583,
      "eval_runtime": 1.0593,
      "eval_samples_per_second": 3.776,
      "eval_steps_per_second": 0.944,
      "step": 1
    },
    {
      "epoch": 0.4444444444444444,
      "grad_norm": 1.828792691230774,
      "learning_rate": 2e-05,
      "loss": 1.3253,
      "step": 2
    },
    {
      "epoch": 0.4444444444444444,
      "eval_loss": 1.5620224475860596,
      "eval_runtime": 0.4406,
      "eval_samples_per_second": 9.078,
      "eval_steps_per_second": 2.27,
      "step": 2
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 1.388032078742981,
      "learning_rate": 3e-05,
      "loss": 1.1657,
      "step": 3
    },
    {
      "epoch": 0.8888888888888888,
      "grad_norm": 1.2471970319747925,
      "learning_rate": 4e-05,
      "loss": 1.4197,
      "step": 4
    },
    {
      "epoch": 0.8888888888888888,
      "eval_loss": 1.5531158447265625,
      "eval_runtime": 0.4432,
      "eval_samples_per_second": 9.025,
      "eval_steps_per_second": 2.256,
      "step": 4
    },
    {
      "epoch": 1.1111111111111112,
      "grad_norm": 1.8095999956130981,
      "learning_rate": 5e-05,
      "loss": 1.983,
      "step": 5
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 1.5582659244537354,
      "learning_rate": 6e-05,
      "loss": 1.6268,
      "step": 6
    },
    {
      "epoch": 1.3333333333333333,
      "eval_loss": 1.5127739906311035,
      "eval_runtime": 0.4404,
      "eval_samples_per_second": 9.083,
      "eval_steps_per_second": 2.271,
      "step": 6
    },
    {
      "epoch": 1.5555555555555556,
      "grad_norm": 1.1906861066818237,
      "learning_rate": 7e-05,
      "loss": 1.1765,
      "step": 7
    },
    {
      "epoch": 1.7777777777777777,
      "grad_norm": 1.2508937120437622,
      "learning_rate": 8e-05,
      "loss": 0.778,
      "step": 8
    },
    {
      "epoch": 1.7777777777777777,
      "eval_loss": 1.4048891067504883,
      "eval_runtime": 0.4436,
      "eval_samples_per_second": 9.018,
      "eval_steps_per_second": 2.254,
      "step": 8
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.9969146251678467,
      "learning_rate": 9e-05,
      "loss": 2.0424,
      "step": 9
    },
    {
      "epoch": 2.2222222222222223,
      "grad_norm": 1.4303289651870728,
      "learning_rate": 0.0001,
      "loss": 1.008,
      "step": 10
    },
    {
      "epoch": 2.2222222222222223,
      "eval_loss": 1.280597448348999,
      "eval_runtime": 0.4402,
      "eval_samples_per_second": 9.087,
      "eval_steps_per_second": 2.272,
      "step": 10
    },
    {
      "epoch": 2.4444444444444446,
      "grad_norm": 1.8639642000198364,
      "learning_rate": 8.535533905932738e-05,
      "loss": 1.1167,
      "step": 11
    },
    {
      "epoch": 2.6666666666666665,
      "grad_norm": 1.3306723833084106,
      "learning_rate": 5e-05,
      "loss": 0.7559,
      "step": 12
    },
    {
      "epoch": 2.6666666666666665,
      "eval_loss": 1.2314685583114624,
      "eval_runtime": 0.4401,
      "eval_samples_per_second": 9.09,
      "eval_steps_per_second": 2.272,
      "step": 12
    },
    {
      "epoch": 2.888888888888889,
      "grad_norm": 1.3251657485961914,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 0.6439,
      "step": 13
    },
    {
      "epoch": 3.111111111111111,
      "grad_norm": 2.698450803756714,
      "learning_rate": 0.0,
      "loss": 1.5859,
      "step": 14
    },
    {
      "epoch": 3.111111111111111,
      "eval_loss": 1.2160450220108032,
      "eval_runtime": 0.4447,
      "eval_samples_per_second": 8.995,
      "eval_steps_per_second": 2.249,
      "step": 14
    }
  ],
  "logging_steps": 1,
  "max_steps": 14,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 25,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.669246388011008e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}