model/checkpoint-36/trainer_state.json

{
  "best_metric": 5.127074241638184,
  "best_model_checkpoint": "./outputs/llama2-13B-lora-QuArch_0_1_1_alpaca_filtered-answer-context-test-new/checkpoint-4",
  "epoch": 7.578947368421053,
  "eval_steps": 4,
  "global_step": 36,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.21052631578947367,
      "eval_accuracy": 0.6153846153846154,
      "eval_loss": 5.132209300994873,
      "eval_runtime": 14.9638,
      "eval_samples_per_second": 3.475,
      "eval_steps_per_second": 1.738,
      "step": 1
    },
    {
      "epoch": 0.8421052631578947,
      "eval_accuracy": 0.6346153846153846,
      "eval_loss": 5.127074241638184,
      "eval_runtime": 15.5565,
      "eval_samples_per_second": 3.343,
      "eval_steps_per_second": 1.671,
      "step": 4
    },
    {
      "epoch": 1.6842105263157894,
      "eval_accuracy": 0.6538461538461539,
      "eval_loss": 5.060066223144531,
      "eval_runtime": 15.7285,
      "eval_samples_per_second": 3.306,
      "eval_steps_per_second": 1.653,
      "step": 8
    },
    {
      "epoch": 2.1052631578947367,
      "grad_norm": 2.453125,
      "learning_rate": 2e-05,
      "loss": 5.1323,
      "step": 10
    },
    {
      "epoch": 2.526315789473684,
      "eval_accuracy": 0.7884615384615384,
      "eval_loss": 4.774318695068359,
      "eval_runtime": 15.4298,
      "eval_samples_per_second": 3.37,
      "eval_steps_per_second": 1.685,
      "step": 12
    },
    {
      "epoch": 3.3684210526315788,
      "eval_accuracy": 0.9230769230769231,
      "eval_loss": 4.04908561706543,
      "eval_runtime": 15.16,
      "eval_samples_per_second": 3.43,
      "eval_steps_per_second": 1.715,
      "step": 16
    },
    {
      "epoch": 4.2105263157894735,
      "grad_norm": 6.03125,
      "learning_rate": 1.8181818181818182e-05,
      "loss": 4.2735,
      "step": 20
    },
    {
      "epoch": 4.2105263157894735,
      "eval_accuracy": 0.8846153846153846,
      "eval_loss": 2.6443593502044678,
      "eval_runtime": 15.6094,
      "eval_samples_per_second": 3.331,
      "eval_steps_per_second": 1.666,
      "step": 20
    },
    {
      "epoch": 5.052631578947368,
      "eval_accuracy": 0.9615384615384616,
      "eval_loss": 1.0550649166107178,
      "eval_runtime": 15.8572,
      "eval_samples_per_second": 3.279,
      "eval_steps_per_second": 1.64,
      "step": 24
    },
    {
      "epoch": 5.894736842105263,
      "eval_accuracy": 0.6923076923076923,
      "eval_loss": 0.4698377847671509,
      "eval_runtime": 15.837,
      "eval_samples_per_second": 3.283,
      "eval_steps_per_second": 1.642,
      "step": 28
    },
    {
      "epoch": 6.315789473684211,
      "grad_norm": 0.6875,
      "learning_rate": 1.6363636363636366e-05,
      "loss": 1.2232,
      "step": 30
    },
    {
      "epoch": 6.7368421052631575,
      "eval_accuracy": 0.6730769230769231,
      "eval_loss": 0.3224477767944336,
      "eval_runtime": 15.7452,
      "eval_samples_per_second": 3.303,
      "eval_steps_per_second": 1.651,
      "step": 32
    },
    {
      "epoch": 7.578947368421053,
      "eval_accuracy": 1.0,
      "eval_loss": 0.2526957392692566,
      "eval_runtime": 15.7721,
      "eval_samples_per_second": 3.297,
      "eval_steps_per_second": 1.648,
      "step": 36
    }
  ],
  "logging_steps": 10,
  "max_steps": 120,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 30,
  "save_steps": 4,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 9.185115697997414e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}