last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.012319063751154912,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0012319063751154912,
      "grad_norm": 5.154358386993408,
      "learning_rate": 2e-05,
      "loss": 6.7099,
      "step": 1
    },
    {
      "epoch": 0.0012319063751154912,
      "eval_loss": 7.131762504577637,
      "eval_runtime": 32.5268,
      "eval_samples_per_second": 10.514,
      "eval_steps_per_second": 5.257,
      "step": 1
    },
    {
      "epoch": 0.0024638127502309825,
      "grad_norm": 4.187411308288574,
      "learning_rate": 4e-05,
      "loss": 7.2327,
      "step": 2
    },
    {
      "epoch": 0.0036957191253464737,
      "grad_norm": 5.173298358917236,
      "learning_rate": 6e-05,
      "loss": 7.0664,
      "step": 3
    },
    {
      "epoch": 0.0036957191253464737,
      "eval_loss": 7.0922064781188965,
      "eval_runtime": 33.2508,
      "eval_samples_per_second": 10.285,
      "eval_steps_per_second": 5.143,
      "step": 3
    },
    {
      "epoch": 0.004927625500461965,
      "grad_norm": 5.8233442306518555,
      "learning_rate": 8e-05,
      "loss": 7.1382,
      "step": 4
    },
    {
      "epoch": 0.006159531875577456,
      "grad_norm": 5.3436808586120605,
      "learning_rate": 0.0001,
      "loss": 6.3826,
      "step": 5
    },
    {
      "epoch": 0.0073914382506929475,
      "grad_norm": 7.44981050491333,
      "learning_rate": 0.00012,
      "loss": 6.336,
      "step": 6
    },
    {
      "epoch": 0.0073914382506929475,
      "eval_loss": 6.42929220199585,
      "eval_runtime": 34.1804,
      "eval_samples_per_second": 10.006,
      "eval_steps_per_second": 5.003,
      "step": 6
    },
    {
      "epoch": 0.008623344625808438,
      "grad_norm": 8.333545684814453,
      "learning_rate": 0.00014,
      "loss": 6.5615,
      "step": 7
    },
    {
      "epoch": 0.00985525100092393,
      "grad_norm": 9.790879249572754,
      "learning_rate": 0.00016,
      "loss": 5.41,
      "step": 8
    },
    {
      "epoch": 0.01108715737603942,
      "grad_norm": 10.339702606201172,
      "learning_rate": 0.00018,
      "loss": 3.5182,
      "step": 9
    },
    {
      "epoch": 0.01108715737603942,
      "eval_loss": 3.122492551803589,
      "eval_runtime": 34.5444,
      "eval_samples_per_second": 9.9,
      "eval_steps_per_second": 4.95,
      "step": 9
    },
    {
      "epoch": 0.012319063751154912,
      "grad_norm": 8.057744979858398,
      "learning_rate": 0.0002,
      "loss": 4.0116,
      "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": 1730214069534720.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}