trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9263157894736842,
  "eval_steps": 3,
  "global_step": 11,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08,
      "grad_norm": 0.3172740936279297,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 0.9286,
      "step": 1
    },
    {
      "epoch": 0.08,
      "eval_loss": 1.0365231037139893,
      "eval_runtime": 24.9628,
      "eval_samples_per_second": 0.801,
      "eval_steps_per_second": 0.801,
      "step": 1
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.3426271378993988,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.951,
      "step": 2
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.28614237904548645,
      "learning_rate": 3e-06,
      "loss": 0.9238,
      "step": 3
    },
    {
      "epoch": 0.25,
      "eval_loss": 1.0386725664138794,
      "eval_runtime": 25.232,
      "eval_samples_per_second": 0.793,
      "eval_steps_per_second": 0.793,
      "step": 3
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.3368183672428131,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.9443,
      "step": 4
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.37111783027648926,
      "learning_rate": 5e-06,
      "loss": 0.9553,
      "step": 5
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.3505602478981018,
      "learning_rate": 6e-06,
      "loss": 0.9448,
      "step": 6
    },
    {
      "epoch": 0.51,
      "eval_loss": 1.0401721000671387,
      "eval_runtime": 25.2324,
      "eval_samples_per_second": 0.793,
      "eval_steps_per_second": 0.793,
      "step": 6
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.27382639050483704,
      "learning_rate": 7e-06,
      "loss": 0.9339,
      "step": 7
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.3769858479499817,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.9662,
      "step": 8
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.3182668387889862,
      "learning_rate": 9e-06,
      "loss": 0.9182,
      "step": 9
    },
    {
      "epoch": 0.76,
      "eval_loss": 1.0365254878997803,
      "eval_runtime": 25.3228,
      "eval_samples_per_second": 0.79,
      "eval_steps_per_second": 0.79,
      "step": 9
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.32636308670043945,
      "learning_rate": 1e-05,
      "loss": 0.959,
      "step": 10
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.2772234082221985,
      "learning_rate": 0.0,
      "loss": 0.9204,
      "step": 11
    }
  ],
  "logging_steps": 1,
  "max_steps": 11,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "total_flos": 1.1152453558861824e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}