last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.045714285714285714,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.004571428571428572,
      "grad_norm": 0.7794408202171326,
      "learning_rate": 2e-05,
      "loss": 2.9356,
      "step": 1
    },
    {
      "epoch": 0.004571428571428572,
      "eval_loss": 2.9138591289520264,
      "eval_runtime": 6.5593,
      "eval_samples_per_second": 14.178,
      "eval_steps_per_second": 7.165,
      "step": 1
    },
    {
      "epoch": 0.009142857142857144,
      "grad_norm": 0.818085253238678,
      "learning_rate": 4e-05,
      "loss": 2.9643,
      "step": 2
    },
    {
      "epoch": 0.013714285714285714,
      "grad_norm": 0.7752296328544617,
      "learning_rate": 6e-05,
      "loss": 3.0035,
      "step": 3
    },
    {
      "epoch": 0.013714285714285714,
      "eval_loss": 2.9076216220855713,
      "eval_runtime": 5.1901,
      "eval_samples_per_second": 17.919,
      "eval_steps_per_second": 9.056,
      "step": 3
    },
    {
      "epoch": 0.018285714285714287,
      "grad_norm": 0.7289955615997314,
      "learning_rate": 8e-05,
      "loss": 2.9317,
      "step": 4
    },
    {
      "epoch": 0.022857142857142857,
      "grad_norm": 0.8006905317306519,
      "learning_rate": 0.0001,
      "loss": 2.8297,
      "step": 5
    },
    {
      "epoch": 0.027428571428571427,
      "grad_norm": 0.9357383847236633,
      "learning_rate": 0.00012,
      "loss": 3.3155,
      "step": 6
    },
    {
      "epoch": 0.027428571428571427,
      "eval_loss": 2.8182923793792725,
      "eval_runtime": 5.1986,
      "eval_samples_per_second": 17.89,
      "eval_steps_per_second": 9.041,
      "step": 6
    },
    {
      "epoch": 0.032,
      "grad_norm": 0.7520653605461121,
      "learning_rate": 0.00014,
      "loss": 2.9988,
      "step": 7
    },
    {
      "epoch": 0.036571428571428574,
      "grad_norm": 0.684337317943573,
      "learning_rate": 0.00016,
      "loss": 2.7895,
      "step": 8
    },
    {
      "epoch": 0.04114285714285714,
      "grad_norm": 0.7094518542289734,
      "learning_rate": 0.00018,
      "loss": 2.9869,
      "step": 9
    },
    {
      "epoch": 0.04114285714285714,
      "eval_loss": 2.6812946796417236,
      "eval_runtime": 5.1981,
      "eval_samples_per_second": 17.891,
      "eval_steps_per_second": 9.042,
      "step": 9
    },
    {
      "epoch": 0.045714285714285714,
      "grad_norm": 0.6055176258087158,
      "learning_rate": 0.0002,
      "loss": 2.6314,
      "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": 3699128496291840.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}