last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.07973421926910298,
  "eval_steps": 3,
  "global_step": 6,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.013289036544850499,
      "grad_norm": 1.4394168853759766,
      "learning_rate": 2e-05,
      "loss": 1.216,
      "step": 1
    },
    {
      "epoch": 0.013289036544850499,
      "eval_loss": 1.447608470916748,
      "eval_runtime": 2.1585,
      "eval_samples_per_second": 14.825,
      "eval_steps_per_second": 7.413,
      "step": 1
    },
    {
      "epoch": 0.026578073089700997,
      "grad_norm": 1.103517770767212,
      "learning_rate": 4e-05,
      "loss": 1.3806,
      "step": 2
    },
    {
      "epoch": 0.03986710963455149,
      "grad_norm": 0.9370229840278625,
      "learning_rate": 6e-05,
      "loss": 1.3254,
      "step": 3
    },
    {
      "epoch": 0.03986710963455149,
      "eval_loss": 1.4428441524505615,
      "eval_runtime": 2.1527,
      "eval_samples_per_second": 14.865,
      "eval_steps_per_second": 7.432,
      "step": 3
    },
    {
      "epoch": 0.053156146179401995,
      "grad_norm": 1.1863428354263306,
      "learning_rate": 8e-05,
      "loss": 1.5754,
      "step": 4
    },
    {
      "epoch": 0.0664451827242525,
      "grad_norm": 1.072548508644104,
      "learning_rate": 0.0001,
      "loss": 1.4127,
      "step": 5
    },
    {
      "epoch": 0.07973421926910298,
      "grad_norm": 0.8001244068145752,
      "learning_rate": 0.00012,
      "loss": 1.3504,
      "step": 6
    },
    {
      "epoch": 0.07973421926910298,
      "eval_loss": 1.3731586933135986,
      "eval_runtime": 2.161,
      "eval_samples_per_second": 14.808,
      "eval_steps_per_second": 7.404,
      "step": 6
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 166605103300608.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}