trainer_state.json

{
  "best_metric": 0.5901639344262295,
  "best_model_checkpoint": "problem2-test/checkpoint-21",
  "epoch": 4.903225806451613,
  "global_step": 35,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.9,
      "eval_accuracy": 0.0,
      "eval_loss": 6.274798393249512,
      "eval_runtime": 1.2157,
      "eval_samples_per_second": 50.175,
      "eval_steps_per_second": 6.58,
      "step": 7
    },
    {
      "epoch": 1.39,
      "learning_rate": 4.032258064516129e-05,
      "loss": 7.4671,
      "step": 10
    },
    {
      "epoch": 1.9,
      "eval_accuracy": 0.5737704918032787,
      "eval_loss": 5.177652835845947,
      "eval_runtime": 1.19,
      "eval_samples_per_second": 51.26,
      "eval_steps_per_second": 6.723,
      "step": 14
    },
    {
      "epoch": 2.77,
      "learning_rate": 2.4193548387096777e-05,
      "loss": 5.3033,
      "step": 20
    },
    {
      "epoch": 2.9,
      "eval_accuracy": 0.5901639344262295,
      "eval_loss": 4.461605072021484,
      "eval_runtime": 1.0892,
      "eval_samples_per_second": 56.006,
      "eval_steps_per_second": 7.345,
      "step": 21
    },
    {
      "epoch": 3.9,
      "eval_accuracy": 0.5245901639344263,
      "eval_loss": 5.181081295013428,
      "eval_runtime": 1.1367,
      "eval_samples_per_second": 53.664,
      "eval_steps_per_second": 7.038,
      "step": 28
    },
    {
      "epoch": 4.26,
      "learning_rate": 8.064516129032258e-06,
      "loss": 4.1105,
      "step": 30
    },
    {
      "epoch": 4.9,
      "eval_accuracy": 0.5573770491803278,
      "eval_loss": 4.228121757507324,
      "eval_runtime": 1.1738,
      "eval_samples_per_second": 51.966,
      "eval_steps_per_second": 6.815,
      "step": 35
    },
    {
      "epoch": 4.9,
      "step": 35,
      "total_flos": 2.76987521728512e+16,
      "train_loss": 5.3129854474748885,
      "train_runtime": 102.3477,
      "train_samples_per_second": 11.92,
      "train_steps_per_second": 0.342
    }
  ],
  "max_steps": 35,
  "num_train_epochs": 5,
  "total_flos": 2.76987521728512e+16,
  "trial_name": null,
  "trial_params": null
}