---
base_model: BitStarWalkin/SuperCorrect-7B
language:
- en
library_name: transformers
license: apache-2.0
metrics:
- accuracy
pipeline_tag: text-generation
tags:
- llama-cpp
- gguf-my-repo
---

# Triangle104/SuperCorrect-7B-Q6_K-GGUF
This model was converted to GGUF format from [`BitStarWalkin/SuperCorrect-7B`](https://huggingface.co./BitStarWalkin/SuperCorrect-7B) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co./spaces/ggml-org/gguf-my-repo) space.
Refer to the [original model card](https://huggingface.co./BitStarWalkin/SuperCorrect-7B) for more details on the model.

---

## Model Description
SuperCorrect is a novel two-stage fine-tuning method for improving both reasoning accuracy and self-correction ability for LLMs. It incorporates LLMs with a pre-defined hierarchical thought template (Buffer of Thought (BoT)) to conduct more deliberate reasoning. Notably, SuperCorrect-7B significantly surpasses powerful DeepSeekMath-7B by 7.8%/5.3% and Qwen2.5-Math-7B by 15.1%/6.3% on MATH/GSM8K benchmarks, achieving new SOTA performance among all 7B models. It relies on pure mathematical reasoning abilities of LLMs, instead of leveraging other programming methods such as PoT and ToRA.

## Paper

[SuperCorrect: Supervising and Correcting Language Models with Error-Driven Insights](https://hf.co/papers/2410.09008)

## Code

https://github.com/YangLing0818/SuperCorrect-llm


## Use with llama.cpp
Install llama.cpp through brew (works on Mac and Linux)

```bash
brew install llama.cpp

```
Invoke the llama.cpp server or the CLI.

### CLI:
```bash
llama-cli --hf-repo Triangle104/SuperCorrect-7B-Q6_K-GGUF --hf-file supercorrect-7b-q6_k.gguf -p "The meaning to life and the universe is"
```

### Server:
```bash
llama-server --hf-repo Triangle104/SuperCorrect-7B-Q6_K-GGUF --hf-file supercorrect-7b-q6_k.gguf -c 2048
```

Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well.

Step 1: Clone llama.cpp from GitHub.
```
git clone https://github.com/ggerganov/llama.cpp
```

Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).
```
cd llama.cpp && LLAMA_CURL=1 make
```

Step 3: Run inference through the main binary.
```
./llama-cli --hf-repo Triangle104/SuperCorrect-7B-Q6_K-GGUF --hf-file supercorrect-7b-q6_k.gguf -p "The meaning to life and the universe is"
```
or 
```
./llama-server --hf-repo Triangle104/SuperCorrect-7B-Q6_K-GGUF --hf-file supercorrect-7b-q6_k.gguf -c 2048
```