last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.7843137254901961,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0784313725490196,
      "grad_norm": 56.80657958984375,
      "learning_rate": 2e-05,
      "loss": 4.5317,
      "step": 1
    },
    {
      "epoch": 0.0784313725490196,
      "eval_loss": 1.4086369276046753,
      "eval_runtime": 0.2104,
      "eval_samples_per_second": 28.52,
      "eval_steps_per_second": 14.26,
      "step": 1
    },
    {
      "epoch": 0.1568627450980392,
      "grad_norm": 65.0270004272461,
      "learning_rate": 4e-05,
      "loss": 5.4882,
      "step": 2
    },
    {
      "epoch": 0.23529411764705882,
      "grad_norm": 56.104896545410156,
      "learning_rate": 6e-05,
      "loss": 6.384,
      "step": 3
    },
    {
      "epoch": 0.23529411764705882,
      "eval_loss": 1.095396876335144,
      "eval_runtime": 0.2099,
      "eval_samples_per_second": 28.581,
      "eval_steps_per_second": 14.291,
      "step": 3
    },
    {
      "epoch": 0.3137254901960784,
      "grad_norm": 61.97636413574219,
      "learning_rate": 8e-05,
      "loss": 4.6142,
      "step": 4
    },
    {
      "epoch": 0.39215686274509803,
      "grad_norm": 37.48371887207031,
      "learning_rate": 0.0001,
      "loss": 2.535,
      "step": 5
    },
    {
      "epoch": 0.47058823529411764,
      "grad_norm": 24.155080795288086,
      "learning_rate": 0.00012,
      "loss": 1.4195,
      "step": 6
    },
    {
      "epoch": 0.47058823529411764,
      "eval_loss": 0.2841545641422272,
      "eval_runtime": 0.208,
      "eval_samples_per_second": 28.853,
      "eval_steps_per_second": 14.426,
      "step": 6
    },
    {
      "epoch": 0.5490196078431373,
      "grad_norm": 30.342317581176758,
      "learning_rate": 0.00014,
      "loss": 1.3593,
      "step": 7
    },
    {
      "epoch": 0.6274509803921569,
      "grad_norm": 14.366281509399414,
      "learning_rate": 0.00016,
      "loss": 0.8921,
      "step": 8
    },
    {
      "epoch": 0.7058823529411765,
      "grad_norm": 13.949260711669922,
      "learning_rate": 0.00018,
      "loss": 0.5549,
      "step": 9
    },
    {
      "epoch": 0.7058823529411765,
      "eval_loss": 0.22411371767520905,
      "eval_runtime": 0.2091,
      "eval_samples_per_second": 28.693,
      "eval_steps_per_second": 14.346,
      "step": 9
    },
    {
      "epoch": 0.7843137254901961,
      "grad_norm": 19.129446029663086,
      "learning_rate": 0.0002,
      "loss": 0.8281,
      "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": 1796486530596864.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}