last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0007731560228854183,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 7.731560228854182e-05,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 7.731560228854182e-05,
      "eval_loss": NaN,
      "eval_runtime": 298.7535,
      "eval_samples_per_second": 18.229,
      "eval_steps_per_second": 9.115,
      "step": 1
    },
    {
      "epoch": 0.00015463120457708364,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 0.0002319468068656255,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.0002319468068656255,
      "eval_loss": NaN,
      "eval_runtime": 297.7561,
      "eval_samples_per_second": 18.29,
      "eval_steps_per_second": 9.145,
      "step": 3
    },
    {
      "epoch": 0.0003092624091541673,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 2.0009,
      "step": 4
    },
    {
      "epoch": 0.00038657801144270914,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.000463893613731251,
      "grad_norm": NaN,
      "learning_rate": 0.00012,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.000463893613731251,
      "eval_loss": NaN,
      "eval_runtime": 298.0089,
      "eval_samples_per_second": 18.275,
      "eval_steps_per_second": 9.137,
      "step": 6
    },
    {
      "epoch": 0.0005412092160197928,
      "grad_norm": NaN,
      "learning_rate": 0.00014,
      "loss": 0.4653,
      "step": 7
    },
    {
      "epoch": 0.0006185248183083346,
      "grad_norm": NaN,
      "learning_rate": 0.00016,
      "loss": 2.4319,
      "step": 8
    },
    {
      "epoch": 0.0006958404205968764,
      "grad_norm": NaN,
      "learning_rate": 0.00018,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.0006958404205968764,
      "eval_loss": NaN,
      "eval_runtime": 297.9934,
      "eval_samples_per_second": 18.276,
      "eval_steps_per_second": 9.138,
      "step": 9
    },
    {
      "epoch": 0.0007731560228854183,
      "grad_norm": NaN,
      "learning_rate": 0.0002,
      "loss": 1.1703,
      "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": 3216747943428096.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}