trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 12.0,
  "eval_steps": 500,
  "global_step": 15,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.8,
      "eval_loss": 2.348168134689331,
      "eval_runtime": 27.1441,
      "eval_samples_per_second": 2.947,
      "eval_steps_per_second": 1.474,
      "step": 1
    },
    {
      "epoch": 1.6,
      "eval_loss": 2.343932628631592,
      "eval_runtime": 27.1355,
      "eval_samples_per_second": 2.948,
      "eval_steps_per_second": 1.474,
      "step": 2
    },
    {
      "epoch": 2.4,
      "eval_loss": 2.338569164276123,
      "eval_runtime": 27.1328,
      "eval_samples_per_second": 2.948,
      "eval_steps_per_second": 1.474,
      "step": 3
    },
    {
      "epoch": 4.0,
      "learning_rate": 1.5000000000000002e-05,
      "loss": 2.3422,
      "step": 5
    },
    {
      "epoch": 4.0,
      "eval_loss": 2.3264243602752686,
      "eval_runtime": 27.1569,
      "eval_samples_per_second": 2.946,
      "eval_steps_per_second": 1.473,
      "step": 5
    },
    {
      "epoch": 4.8,
      "eval_loss": 2.321211576461792,
      "eval_runtime": 27.1325,
      "eval_samples_per_second": 2.948,
      "eval_steps_per_second": 1.474,
      "step": 6
    },
    {
      "epoch": 5.6,
      "eval_loss": 2.3165886402130127,
      "eval_runtime": 27.1266,
      "eval_samples_per_second": 2.949,
      "eval_steps_per_second": 1.475,
      "step": 7
    },
    {
      "epoch": 6.4,
      "eval_loss": 2.3125572204589844,
      "eval_runtime": 27.13,
      "eval_samples_per_second": 2.949,
      "eval_steps_per_second": 1.474,
      "step": 8
    },
    {
      "epoch": 8.0,
      "learning_rate": 5.000000000000003e-06,
      "loss": 2.3191,
      "step": 10
    },
    {
      "epoch": 8.0,
      "eval_loss": 2.307173252105713,
      "eval_runtime": 27.1321,
      "eval_samples_per_second": 2.949,
      "eval_steps_per_second": 1.474,
      "step": 10
    },
    {
      "epoch": 8.8,
      "eval_loss": 2.3055243492126465,
      "eval_runtime": 27.1317,
      "eval_samples_per_second": 2.949,
      "eval_steps_per_second": 1.474,
      "step": 11
    },
    {
      "epoch": 9.6,
      "eval_loss": 2.3045384883880615,
      "eval_runtime": 27.1304,
      "eval_samples_per_second": 2.949,
      "eval_steps_per_second": 1.474,
      "step": 12
    },
    {
      "epoch": 10.4,
      "eval_loss": 2.3039393424987793,
      "eval_runtime": 27.1363,
      "eval_samples_per_second": 2.948,
      "eval_steps_per_second": 1.474,
      "step": 13
    },
    {
      "epoch": 12.0,
      "learning_rate": 0.0,
      "loss": 2.3082,
      "step": 15
    },
    {
      "epoch": 12.0,
      "eval_loss": 2.3035786151885986,
      "eval_runtime": 27.1238,
      "eval_samples_per_second": 2.949,
      "eval_steps_per_second": 1.475,
      "step": 15
    },
    {
      "epoch": 12.0,
      "step": 15,
      "total_flos": 3.3809589564604416e+17,
      "train_loss": 2.323172187805176,
      "train_runtime": 4255.0402,
      "train_samples_per_second": 1.128,
      "train_steps_per_second": 0.004
    }
  ],
  "logging_steps": 5,
  "max_steps": 15,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 15,
  "save_steps": 500,
  "total_flos": 3.3809589564604416e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}