last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00018811490058127505,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.8811490058127504e-05,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 23.1669,
      "step": 1
    },
    {
      "epoch": 1.8811490058127504e-05,
      "eval_loss": 5.918900489807129,
      "eval_runtime": 2202.7103,
      "eval_samples_per_second": 5.081,
      "eval_steps_per_second": 5.081,
      "step": 1
    },
    {
      "epoch": 3.762298011625501e-05,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 27.6838,
      "step": 2
    },
    {
      "epoch": 5.6434470174382516e-05,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 13.1598,
      "step": 3
    },
    {
      "epoch": 5.6434470174382516e-05,
      "eval_loss": 5.918900489807129,
      "eval_runtime": 2202.0869,
      "eval_samples_per_second": 5.082,
      "eval_steps_per_second": 5.082,
      "step": 3
    },
    {
      "epoch": 7.524596023251002e-05,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 20.3679,
      "step": 4
    },
    {
      "epoch": 9.405745029063752e-05,
      "grad_norm": 88.73544311523438,
      "learning_rate": 2e-05,
      "loss": 29.1746,
      "step": 5
    },
    {
      "epoch": 0.00011286894034876503,
      "grad_norm": 71.97441864013672,
      "learning_rate": 4e-05,
      "loss": 21.5281,
      "step": 6
    },
    {
      "epoch": 0.00011286894034876503,
      "eval_loss": 5.739979267120361,
      "eval_runtime": 2197.5616,
      "eval_samples_per_second": 5.093,
      "eval_steps_per_second": 5.093,
      "step": 6
    },
    {
      "epoch": 0.00013168043040689254,
      "grad_norm": 92.83966064453125,
      "learning_rate": 6e-05,
      "loss": 22.5883,
      "step": 7
    },
    {
      "epoch": 0.00015049192046502003,
      "grad_norm": 93.7508773803711,
      "learning_rate": 8e-05,
      "loss": 22.5171,
      "step": 8
    },
    {
      "epoch": 0.00016930341052314753,
      "grad_norm": 114.79270935058594,
      "learning_rate": 0.0001,
      "loss": 17.7258,
      "step": 9
    },
    {
      "epoch": 0.00016930341052314753,
      "eval_loss": 3.5046584606170654,
      "eval_runtime": 2197.756,
      "eval_samples_per_second": 5.092,
      "eval_steps_per_second": 5.092,
      "step": 9
    },
    {
      "epoch": 0.00018811490058127505,
      "grad_norm": 92.64408111572266,
      "learning_rate": 0.00012,
      "loss": 11.9718,
      "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": 7031294986813440.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}