last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.005533484676503973,
  "eval_steps": 10,
  "global_step": 39,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00014188422247446084,
      "eval_loss": 0.9373186230659485,
      "eval_runtime": 85.4442,
      "eval_samples_per_second": 34.736,
      "eval_steps_per_second": 17.368,
      "step": 1
    },
    {
      "epoch": 0.0007094211123723042,
      "grad_norm": 0.4906405508518219,
      "learning_rate": 5e-05,
      "loss": 0.9687,
      "step": 5
    },
    {
      "epoch": 0.0014188422247446084,
      "grad_norm": 0.6939789056777954,
      "learning_rate": 0.0001,
      "loss": 0.8927,
      "step": 10
    },
    {
      "epoch": 0.0014188422247446084,
      "eval_loss": 0.8500236868858337,
      "eval_runtime": 86.1579,
      "eval_samples_per_second": 34.448,
      "eval_steps_per_second": 17.224,
      "step": 10
    },
    {
      "epoch": 0.0021282633371169125,
      "grad_norm": 0.6280760765075684,
      "learning_rate": 9.619397662556435e-05,
      "loss": 0.7719,
      "step": 15
    },
    {
      "epoch": 0.0028376844494892167,
      "grad_norm": 1.1123079061508179,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.7106,
      "step": 20
    },
    {
      "epoch": 0.0028376844494892167,
      "eval_loss": 0.6546657085418701,
      "eval_runtime": 85.8004,
      "eval_samples_per_second": 34.592,
      "eval_steps_per_second": 17.296,
      "step": 20
    },
    {
      "epoch": 0.003547105561861521,
      "grad_norm": 0.8203548192977905,
      "learning_rate": 6.91341716182545e-05,
      "loss": 0.6668,
      "step": 25
    },
    {
      "epoch": 0.004256526674233825,
      "grad_norm": 0.8009120225906372,
      "learning_rate": 5e-05,
      "loss": 0.6256,
      "step": 30
    },
    {
      "epoch": 0.004256526674233825,
      "eval_loss": 0.6064170002937317,
      "eval_runtime": 86.325,
      "eval_samples_per_second": 34.382,
      "eval_steps_per_second": 17.191,
      "step": 30
    },
    {
      "epoch": 0.004965947786606129,
      "grad_norm": 0.7089608311653137,
      "learning_rate": 3.086582838174551e-05,
      "loss": 0.6629,
      "step": 35
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1028332078497792.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}