last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.036596523330283626,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0036596523330283625,
      "grad_norm": 19.45615005493164,
      "learning_rate": 2e-05,
      "loss": 47.7593,
      "step": 1
    },
    {
      "epoch": 0.0036596523330283625,
      "eval_loss": 11.760443687438965,
      "eval_runtime": 2.0877,
      "eval_samples_per_second": 55.084,
      "eval_steps_per_second": 27.781,
      "step": 1
    },
    {
      "epoch": 0.007319304666056725,
      "grad_norm": 18.177263259887695,
      "learning_rate": 4e-05,
      "loss": 48.1966,
      "step": 2
    },
    {
      "epoch": 0.010978956999085087,
      "grad_norm": 17.849220275878906,
      "learning_rate": 6e-05,
      "loss": 47.3488,
      "step": 3
    },
    {
      "epoch": 0.010978956999085087,
      "eval_loss": 11.744895935058594,
      "eval_runtime": 2.065,
      "eval_samples_per_second": 55.69,
      "eval_steps_per_second": 28.087,
      "step": 3
    },
    {
      "epoch": 0.01463860933211345,
      "grad_norm": 18.0354061126709,
      "learning_rate": 8e-05,
      "loss": 47.6116,
      "step": 4
    },
    {
      "epoch": 0.018298261665141813,
      "grad_norm": 21.11904525756836,
      "learning_rate": 0.0001,
      "loss": 49.3131,
      "step": 5
    },
    {
      "epoch": 0.021957913998170174,
      "grad_norm": 23.17920684814453,
      "learning_rate": 0.00012,
      "loss": 49.1428,
      "step": 6
    },
    {
      "epoch": 0.021957913998170174,
      "eval_loss": 11.213788032531738,
      "eval_runtime": 2.0751,
      "eval_samples_per_second": 55.418,
      "eval_steps_per_second": 27.95,
      "step": 6
    },
    {
      "epoch": 0.025617566331198535,
      "grad_norm": 26.00096893310547,
      "learning_rate": 0.00014,
      "loss": 42.9896,
      "step": 7
    },
    {
      "epoch": 0.0292772186642269,
      "grad_norm": 29.249248504638672,
      "learning_rate": 0.00016,
      "loss": 41.6571,
      "step": 8
    },
    {
      "epoch": 0.03293687099725526,
      "grad_norm": 33.103111267089844,
      "learning_rate": 0.00018,
      "loss": 37.9857,
      "step": 9
    },
    {
      "epoch": 0.03293687099725526,
      "eval_loss": 8.02658748626709,
      "eval_runtime": 2.09,
      "eval_samples_per_second": 55.025,
      "eval_steps_per_second": 27.752,
      "step": 9
    },
    {
      "epoch": 0.036596523330283626,
      "grad_norm": 36.67914962768555,
      "learning_rate": 0.0002,
      "loss": 34.1644,
      "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": 298729403842560.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}