{
  "best_global_step": 30,
  "best_metric": 0.34565383195877075,
  "best_model_checkpoint": "finetuned_models/runpod-qwen-32b-3epochs_202512121522/checkpoint-30",
  "epoch": 2.142857142857143,
  "eval_steps": 5,
  "global_step": 30,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.35714285714285715,
      "eval_loss": 9.715937614440918,
      "eval_runtime": 80.6928,
      "eval_samples_per_second": 0.459,
      "eval_steps_per_second": 0.459,
      "step": 5
    },
    {
      "epoch": 0.7142857142857143,
      "eval_loss": 0.6401041150093079,
      "eval_runtime": 80.8795,
      "eval_samples_per_second": 0.457,
      "eval_steps_per_second": 0.457,
      "step": 10
    },
    {
      "epoch": 1.0714285714285714,
      "eval_loss": 0.4123309552669525,
      "eval_runtime": 80.9572,
      "eval_samples_per_second": 0.457,
      "eval_steps_per_second": 0.457,
      "step": 15
    },
    {
      "epoch": 1.4285714285714286,
      "grad_norm": 0.1379181295633316,
      "learning_rate": 0.00011794871794871796,
      "loss": 5.1548,
      "step": 20
    },
    {
      "epoch": 1.4285714285714286,
      "eval_loss": 0.37876370549201965,
      "eval_runtime": 80.8925,
      "eval_samples_per_second": 0.457,
      "eval_steps_per_second": 0.457,
      "step": 20
    },
    {
      "epoch": 1.7857142857142856,
      "eval_loss": 0.3591781258583069,
      "eval_runtime": 80.7089,
      "eval_samples_per_second": 0.458,
      "eval_steps_per_second": 0.458,
      "step": 25
    },
    {
      "epoch": 2.142857142857143,
      "eval_loss": 0.34565383195877075,
      "eval_runtime": 80.5729,
      "eval_samples_per_second": 0.459,
      "eval_steps_per_second": 0.459,
      "step": 30
    }
  ],
  "logging_steps": 20,
  "max_steps": 42,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.4395433047636378e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}