{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.007497656982193065,
  "eval_steps": 10,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00014995313964386129,
      "eval_loss": 1.9529391527175903,
      "eval_runtime": 331.1234,
      "eval_samples_per_second": 8.48,
      "eval_steps_per_second": 4.24,
      "step": 1
    },
    {
      "epoch": 0.0007497656982193065,
      "grad_norm": 10.662879943847656,
      "learning_rate": 5e-05,
      "loss": 7.8251,
      "step": 5
    },
    {
      "epoch": 0.001499531396438613,
      "grad_norm": 10.276589393615723,
      "learning_rate": 0.0001,
      "loss": 7.0588,
      "step": 10
    },
    {
      "epoch": 0.001499531396438613,
      "eval_loss": 1.8228198289871216,
      "eval_runtime": 332.2916,
      "eval_samples_per_second": 8.45,
      "eval_steps_per_second": 4.225,
      "step": 10
    },
    {
      "epoch": 0.0022492970946579195,
      "grad_norm": 13.335864067077637,
      "learning_rate": 9.619397662556435e-05,
      "loss": 6.9452,
      "step": 15
    },
    {
      "epoch": 0.002999062792877226,
      "grad_norm": 12.599164009094238,
      "learning_rate": 8.535533905932738e-05,
      "loss": 7.4634,
      "step": 20
    },
    {
      "epoch": 0.002999062792877226,
      "eval_loss": 1.7322559356689453,
      "eval_runtime": 332.4149,
      "eval_samples_per_second": 8.447,
      "eval_steps_per_second": 4.224,
      "step": 20
    },
    {
      "epoch": 0.0037488284910965324,
      "grad_norm": 11.365962028503418,
      "learning_rate": 6.91341716182545e-05,
      "loss": 6.8846,
      "step": 25
    },
    {
      "epoch": 0.004498594189315839,
      "grad_norm": 12.568791389465332,
      "learning_rate": 5e-05,
      "loss": 7.0615,
      "step": 30
    },
    {
      "epoch": 0.004498594189315839,
      "eval_loss": 1.6993041038513184,
      "eval_runtime": 332.2798,
      "eval_samples_per_second": 8.451,
      "eval_steps_per_second": 4.225,
      "step": 30
    },
    {
      "epoch": 0.005248359887535145,
      "grad_norm": 9.812939643859863,
      "learning_rate": 3.086582838174551e-05,
      "loss": 6.8076,
      "step": 35
    },
    {
      "epoch": 0.005998125585754452,
      "grad_norm": 10.162247657775879,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 6.6806,
      "step": 40
    },
    {
      "epoch": 0.005998125585754452,
      "eval_loss": 1.6791855096817017,
      "eval_runtime": 333.3267,
      "eval_samples_per_second": 8.424,
      "eval_steps_per_second": 4.212,
      "step": 40
    },
    {
      "epoch": 0.006747891283973758,
      "grad_norm": 11.431609153747559,
      "learning_rate": 3.8060233744356633e-06,
      "loss": 6.7861,
      "step": 45
    },
    {
      "epoch": 0.007497656982193065,
      "grad_norm": 10.58781909942627,
      "learning_rate": 0.0,
      "loss": 6.319,
      "step": 50
    },
    {
      "epoch": 0.007497656982193065,
      "eval_loss": 1.6756504774093628,
      "eval_runtime": 332.7507,
      "eval_samples_per_second": 8.439,
      "eval_steps_per_second": 4.219,
      "step": 50
    }
  ],
  "logging_steps": 5,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 13,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.7482881114832896e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}