last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.8333333333333334,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08333333333333333,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 1.4927,
      "step": 1
    },
    {
      "epoch": 0.08333333333333333,
      "eval_loss": 1.4977792501449585,
      "eval_runtime": 0.453,
      "eval_samples_per_second": 6.622,
      "eval_steps_per_second": 6.622,
      "step": 1
    },
    {
      "epoch": 0.16666666666666666,
      "grad_norm": 4.665465831756592,
      "learning_rate": 2e-05,
      "loss": 1.307,
      "step": 2
    },
    {
      "epoch": 0.25,
      "grad_norm": 5.513491153717041,
      "learning_rate": 4e-05,
      "loss": 1.5735,
      "step": 3
    },
    {
      "epoch": 0.25,
      "eval_loss": 1.4473786354064941,
      "eval_runtime": 0.4679,
      "eval_samples_per_second": 6.412,
      "eval_steps_per_second": 6.412,
      "step": 3
    },
    {
      "epoch": 0.3333333333333333,
      "grad_norm": 4.566956520080566,
      "learning_rate": 6e-05,
      "loss": 1.282,
      "step": 4
    },
    {
      "epoch": 0.4166666666666667,
      "grad_norm": 3.311328649520874,
      "learning_rate": 8e-05,
      "loss": 1.3071,
      "step": 5
    },
    {
      "epoch": 0.5,
      "grad_norm": 2.798964262008667,
      "learning_rate": 0.0001,
      "loss": 1.0892,
      "step": 6
    },
    {
      "epoch": 0.5,
      "eval_loss": 0.8217151761054993,
      "eval_runtime": 0.4475,
      "eval_samples_per_second": 6.705,
      "eval_steps_per_second": 6.705,
      "step": 6
    },
    {
      "epoch": 0.5833333333333334,
      "grad_norm": 2.373387575149536,
      "learning_rate": 0.00012,
      "loss": 0.8337,
      "step": 7
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 2.598729133605957,
      "learning_rate": 0.00014,
      "loss": 0.5417,
      "step": 8
    },
    {
      "epoch": 0.75,
      "grad_norm": 2.8108651638031006,
      "learning_rate": 0.00016,
      "loss": 0.2385,
      "step": 9
    },
    {
      "epoch": 0.75,
      "eval_loss": 0.05185846984386444,
      "eval_runtime": 0.4321,
      "eval_samples_per_second": 6.943,
      "eval_steps_per_second": 6.943,
      "step": 9
    },
    {
      "epoch": 0.8333333333333334,
      "grad_norm": 2.6426496505737305,
      "learning_rate": 0.00018,
      "loss": 0.1522,
      "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": 1849564248145920.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}