last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.008592910848549946,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0008592910848549946,
      "grad_norm": 0.6812443733215332,
      "learning_rate": 2e-05,
      "loss": 1.8317,
      "step": 1
    },
    {
      "epoch": 0.0008592910848549946,
      "eval_loss": 1.6235483884811401,
      "eval_runtime": 29.9071,
      "eval_samples_per_second": 16.384,
      "eval_steps_per_second": 8.192,
      "step": 1
    },
    {
      "epoch": 0.0017185821697099893,
      "grad_norm": 0.5641345381736755,
      "learning_rate": 4e-05,
      "loss": 1.6288,
      "step": 2
    },
    {
      "epoch": 0.0025778732545649837,
      "grad_norm": 0.6060267686843872,
      "learning_rate": 6e-05,
      "loss": 1.6476,
      "step": 3
    },
    {
      "epoch": 0.0025778732545649837,
      "eval_loss": 1.6178789138793945,
      "eval_runtime": 29.3207,
      "eval_samples_per_second": 16.712,
      "eval_steps_per_second": 8.356,
      "step": 3
    },
    {
      "epoch": 0.0034371643394199786,
      "grad_norm": 0.5128967761993408,
      "learning_rate": 8e-05,
      "loss": 1.5166,
      "step": 4
    },
    {
      "epoch": 0.004296455424274973,
      "grad_norm": 0.6520130038261414,
      "learning_rate": 0.0001,
      "loss": 1.5365,
      "step": 5
    },
    {
      "epoch": 0.0051557465091299675,
      "grad_norm": 0.6379581689834595,
      "learning_rate": 0.00012,
      "loss": 1.5424,
      "step": 6
    },
    {
      "epoch": 0.0051557465091299675,
      "eval_loss": 1.5232118368148804,
      "eval_runtime": 29.3237,
      "eval_samples_per_second": 16.71,
      "eval_steps_per_second": 8.355,
      "step": 6
    },
    {
      "epoch": 0.006015037593984963,
      "grad_norm": 0.7008845806121826,
      "learning_rate": 0.00014,
      "loss": 1.4024,
      "step": 7
    },
    {
      "epoch": 0.006874328678839957,
      "grad_norm": 0.7331475615501404,
      "learning_rate": 0.00016,
      "loss": 1.4373,
      "step": 8
    },
    {
      "epoch": 0.007733619763694952,
      "grad_norm": 0.7993574738502502,
      "learning_rate": 0.00018,
      "loss": 1.3813,
      "step": 9
    },
    {
      "epoch": 0.007733619763694952,
      "eval_loss": 1.4099993705749512,
      "eval_runtime": 29.3679,
      "eval_samples_per_second": 16.685,
      "eval_steps_per_second": 8.342,
      "step": 9
    },
    {
      "epoch": 0.008592910848549946,
      "grad_norm": 0.9581007957458496,
      "learning_rate": 0.0002,
      "loss": 1.4072,
      "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": 3606650283884544.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}