bermaneh/pde-mc-logprob-results-v3-deepseek

---
license: mit
tags:
- pde-mc-logprob
- mc
- v3
- deepseek
---

# pde-mc-logprob-results-v3-deepseek

DeepSeek-R1-Distill-Qwen-32B MC text-extraction on v3 dataset (128 rows, 8 conditions, 9 question types).

## Dataset Info

- **Rows**: 1152
- **Columns**: 19

## Columns

| Column | Type | Description |
|--------|------|-------------|
| title | Value('string') | *No description provided* |
| pde_class | Value('string') | *No description provided* |
| mod_type | Value('string') | *No description provided* |
| model | Value('string') | *No description provided* |
| question_type | Value('string') | *No description provided* |
| candidate | Value('string') | *No description provided* |
| letters | Value('string') | *No description provided* |
| correct_letter | Value('string') | *No description provided* |
| logprob_A | Value('null') | *No description provided* |
| logprob_B | Value('null') | *No description provided* |
| logprob_C | Value('null') | *No description provided* |
| logprob_D | Value('null') | *No description provided* |
| predicted_letter | Value('string') | *No description provided* |
| correct | Value('bool') | *No description provided* |
| logprob_correct | Value('null') | *No description provided* |
| margin | Value('null') | *No description provided* |
| entropy | Value('null') | *No description provided* |
| finish_reason | Value('string') | *No description provided* |
| scoring_method | Value('string') | *No description provided* |

## Generation Parameters

```json
{
"script_name": "run_mc_deepseek_only.sbatch",
"model": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
"description": "DeepSeek-R1-Distill-Qwen-32B MC text-extraction on v3 dataset (128 rows, 8 conditions, 9 question types).",
"experiment_name": "pde-mc-logprob",
"job_id": "torch:7256317",
"cluster": "torch",
"artifact_status": "final",
"canary": false,
"hyperparameters": {},
"input_datasets": []
}
```

## Usage

```python
from datasets import load_dataset

dataset = load_dataset("bermaneh/pde-mc-logprob-results-v3-deepseek", split="train")
print(f"Loaded {len(dataset)} rows")
```

---