last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0028146989835809225,
  "eval_steps": 3,
  "global_step": 9,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0003127443315089914,
      "grad_norm": 6.8922343254089355,
      "learning_rate": 2e-05,
      "loss": 19.7839,
      "step": 1
    },
    {
      "epoch": 0.0003127443315089914,
      "eval_loss": 17.957256317138672,
      "eval_runtime": 38.7574,
      "eval_samples_per_second": 34.755,
      "eval_steps_per_second": 17.39,
      "step": 1
    },
    {
      "epoch": 0.0006254886630179827,
      "grad_norm": 6.97463846206665,
      "learning_rate": 4e-05,
      "loss": 16.3235,
      "step": 2
    },
    {
      "epoch": 0.0009382329945269742,
      "grad_norm": 9.537607192993164,
      "learning_rate": 6e-05,
      "loss": 16.7324,
      "step": 3
    },
    {
      "epoch": 0.0009382329945269742,
      "eval_loss": 17.92792320251465,
      "eval_runtime": 37.2858,
      "eval_samples_per_second": 36.126,
      "eval_steps_per_second": 18.077,
      "step": 3
    },
    {
      "epoch": 0.0012509773260359655,
      "grad_norm": 6.914788722991943,
      "learning_rate": 8e-05,
      "loss": 18.15,
      "step": 4
    },
    {
      "epoch": 0.001563721657544957,
      "grad_norm": 8.019819259643555,
      "learning_rate": 0.0001,
      "loss": 16.1831,
      "step": 5
    },
    {
      "epoch": 0.0018764659890539485,
      "grad_norm": 7.5083723068237305,
      "learning_rate": 0.00012,
      "loss": 18.0949,
      "step": 6
    },
    {
      "epoch": 0.0018764659890539485,
      "eval_loss": 17.07736587524414,
      "eval_runtime": 37.4363,
      "eval_samples_per_second": 35.981,
      "eval_steps_per_second": 18.004,
      "step": 6
    },
    {
      "epoch": 0.00218921032056294,
      "grad_norm": 9.1001558303833,
      "learning_rate": 0.00014,
      "loss": 17.5335,
      "step": 7
    },
    {
      "epoch": 0.002501954652071931,
      "grad_norm": 15.724388122558594,
      "learning_rate": 0.00016,
      "loss": 15.0517,
      "step": 8
    },
    {
      "epoch": 0.0028146989835809225,
      "grad_norm": 15.608181953430176,
      "learning_rate": 0.00018,
      "loss": 13.1964,
      "step": 9
    },
    {
      "epoch": 0.0028146989835809225,
      "eval_loss": 13.878823280334473,
      "eval_runtime": 37.4374,
      "eval_samples_per_second": 35.98,
      "eval_steps_per_second": 18.003,
      "step": 9
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1568372441481216.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}