last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.04484304932735426,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.004484304932735426,
      "grad_norm": 0.006755814887583256,
      "learning_rate": 2e-05,
      "loss": 11.9309,
      "step": 1
    },
    {
      "epoch": 0.004484304932735426,
      "eval_loss": 11.931302070617676,
      "eval_runtime": 0.5278,
      "eval_samples_per_second": 1422.827,
      "eval_steps_per_second": 89.045,
      "step": 1
    },
    {
      "epoch": 0.008968609865470852,
      "grad_norm": 0.008150096982717514,
      "learning_rate": 4e-05,
      "loss": 11.9311,
      "step": 2
    },
    {
      "epoch": 0.013452914798206279,
      "grad_norm": 0.00839826837182045,
      "learning_rate": 6e-05,
      "loss": 11.9313,
      "step": 3
    },
    {
      "epoch": 0.013452914798206279,
      "eval_loss": 11.931297302246094,
      "eval_runtime": 0.521,
      "eval_samples_per_second": 1441.543,
      "eval_steps_per_second": 90.216,
      "step": 3
    },
    {
      "epoch": 0.017937219730941704,
      "grad_norm": 0.007786044850945473,
      "learning_rate": 8e-05,
      "loss": 11.9324,
      "step": 4
    },
    {
      "epoch": 0.02242152466367713,
      "grad_norm": 0.009498490951955318,
      "learning_rate": 0.0001,
      "loss": 11.9317,
      "step": 5
    },
    {
      "epoch": 0.026905829596412557,
      "grad_norm": 0.008360755629837513,
      "learning_rate": 0.00012,
      "loss": 11.9318,
      "step": 6
    },
    {
      "epoch": 0.026905829596412557,
      "eval_loss": 11.931273460388184,
      "eval_runtime": 0.5165,
      "eval_samples_per_second": 1454.119,
      "eval_steps_per_second": 91.003,
      "step": 6
    },
    {
      "epoch": 0.03139013452914798,
      "grad_norm": 0.008624019101262093,
      "learning_rate": 0.00014,
      "loss": 11.9322,
      "step": 7
    },
    {
      "epoch": 0.03587443946188341,
      "grad_norm": 0.007526622619479895,
      "learning_rate": 0.00016,
      "loss": 11.9326,
      "step": 8
    },
    {
      "epoch": 0.04035874439461883,
      "grad_norm": 0.006862244568765163,
      "learning_rate": 0.00018,
      "loss": 11.9313,
      "step": 9
    },
    {
      "epoch": 0.04035874439461883,
      "eval_loss": 11.931229591369629,
      "eval_runtime": 0.5325,
      "eval_samples_per_second": 1410.197,
      "eval_steps_per_second": 88.255,
      "step": 9
    },
    {
      "epoch": 0.04484304932735426,
      "grad_norm": 0.007552538998425007,
      "learning_rate": 0.0002,
      "loss": 11.9309,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 9748316160.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}