trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9846153846153847,
  "eval_steps": 500,
  "global_step": 48,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04,
      "learning_rate": 2e-05,
      "loss": 1.8569,
      "step": 2
    },
    {
      "epoch": 0.08,
      "learning_rate": 2e-05,
      "loss": 2.2234,
      "step": 4
    },
    {
      "epoch": 0.12,
      "learning_rate": 2e-05,
      "loss": 2.2528,
      "step": 6
    },
    {
      "epoch": 0.16,
      "learning_rate": 2e-05,
      "loss": 2.3016,
      "step": 8
    },
    {
      "epoch": 0.21,
      "learning_rate": 2e-05,
      "loss": 2.5323,
      "step": 10
    },
    {
      "epoch": 0.25,
      "learning_rate": 2e-05,
      "loss": 3.003,
      "step": 12
    },
    {
      "epoch": 0.29,
      "learning_rate": 2e-05,
      "loss": 1.9097,
      "step": 14
    },
    {
      "epoch": 0.33,
      "learning_rate": 2e-05,
      "loss": 1.8799,
      "step": 16
    },
    {
      "epoch": 0.37,
      "learning_rate": 2e-05,
      "loss": 2.1973,
      "step": 18
    },
    {
      "epoch": 0.41,
      "learning_rate": 2e-05,
      "loss": 2.1347,
      "step": 20
    },
    {
      "epoch": 0.45,
      "learning_rate": 2e-05,
      "loss": 2.4056,
      "step": 22
    },
    {
      "epoch": 0.49,
      "learning_rate": 2e-05,
      "loss": 2.7531,
      "step": 24
    },
    {
      "epoch": 0.53,
      "learning_rate": 2e-05,
      "loss": 1.819,
      "step": 26
    },
    {
      "epoch": 0.57,
      "learning_rate": 2e-05,
      "loss": 2.0862,
      "step": 28
    },
    {
      "epoch": 0.62,
      "learning_rate": 2e-05,
      "loss": 2.3401,
      "step": 30
    },
    {
      "epoch": 0.66,
      "learning_rate": 2e-05,
      "loss": 2.1033,
      "step": 32
    },
    {
      "epoch": 0.7,
      "learning_rate": 2e-05,
      "loss": 2.1674,
      "step": 34
    },
    {
      "epoch": 0.74,
      "learning_rate": 2e-05,
      "loss": 2.7192,
      "step": 36
    },
    {
      "epoch": 0.78,
      "learning_rate": 2e-05,
      "loss": 1.8597,
      "step": 38
    },
    {
      "epoch": 0.82,
      "learning_rate": 2e-05,
      "loss": 2.1725,
      "step": 40
    },
    {
      "epoch": 0.86,
      "learning_rate": 2e-05,
      "loss": 2.2864,
      "step": 42
    },
    {
      "epoch": 0.9,
      "learning_rate": 2e-05,
      "loss": 2.3444,
      "step": 44
    },
    {
      "epoch": 0.94,
      "learning_rate": 2e-05,
      "loss": 2.1558,
      "step": 46
    },
    {
      "epoch": 0.98,
      "learning_rate": 2e-05,
      "loss": 2.661,
      "step": 48
    }
  ],
  "logging_steps": 2,
  "max_steps": 48,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "total_flos": 6711085690343424.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}