trainer_state.json

{
  "best_metric": 0.2744598984718323,
  "best_model_checkpoint": "./ryan_model314_3/checkpoint-100",
  "epoch": 2.0,
  "eval_steps": 50,
  "global_step": 126,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.79,
      "grad_norm": 0.808198094367981,
      "learning_rate": 0.00012063492063492063,
      "loss": 0.4094,
      "step": 50
    },
    {
      "epoch": 0.79,
      "eval_loss": 0.31280139088630676,
      "eval_na_accuracy": 0.94,
      "eval_ordinal_accuracy": 0.40217391304347827,
      "eval_ordinal_mae": 0.6332903986837229,
      "eval_runtime": 11.6495,
      "eval_samples_per_second": 8.584,
      "eval_steps_per_second": 1.116,
      "step": 50
    },
    {
      "epoch": 1.59,
      "grad_norm": 0.8316417336463928,
      "learning_rate": 4.126984126984127e-05,
      "loss": 0.29,
      "step": 100
    },
    {
      "epoch": 1.59,
      "eval_loss": 0.2744598984718323,
      "eval_na_accuracy": 0.96,
      "eval_ordinal_accuracy": 0.5108695652173914,
      "eval_ordinal_mae": 0.5806456123182611,
      "eval_runtime": 4.1502,
      "eval_samples_per_second": 24.095,
      "eval_steps_per_second": 3.132,
      "step": 100
    },
    {
      "epoch": 2.0,
      "step": 126,
      "total_flos": 1.5498953551872e+17,
      "train_loss": 0.32528334193759495,
      "train_runtime": 229.1686,
      "train_samples_per_second": 8.727,
      "train_steps_per_second": 0.55
    }
  ],
  "logging_steps": 50,
  "max_steps": 126,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 50,
  "total_flos": 1.5498953551872e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}