ajibawa-2023_-_Code-Mistral-7B-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

Code-Mistral-7B - GGUF

• Model creator: https://huggingface.co/ajibawa-2023/
• Original model: https://huggingface.co/ajibawa-2023/Code-Mistral-7B/

Original model description:

language:

• en license: apache-2.0 tags:
• code
• mathematics datasets:
• ajibawa-2023/Code-290k-ShareGPT
• m-a-p/Code-Feedback
• microsoft/orca-math-word-problems-200k
• teknium/openhermes model-index:
• name: Code-Mistral-7B results:
  • task: type: text-generation name: Text Generation dataset: name: AI2 Reasoning Challenge (25-Shot) type: ai2_arc config: ARC-Challenge split: test args: num_few_shot: 25 metrics:
    • type: acc_norm value: 64.59 name: normalized accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=ajibawa-2023/Code-Mistral-7B name: Open LLM Leaderboard
  • task: type: text-generation name: Text Generation dataset: name: HellaSwag (10-Shot) type: hellaswag split: validation args: num_few_shot: 10 metrics:
    • type: acc_norm value: 85.29 name: normalized accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=ajibawa-2023/Code-Mistral-7B name: Open LLM Leaderboard
  • task: type: text-generation name: Text Generation dataset: name: MMLU (5-Shot) type: cais/mmlu config: all split: test args: num_few_shot: 5 metrics:
    • type: acc value: 65.0 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=ajibawa-2023/Code-Mistral-7B name: Open LLM Leaderboard
  • task: type: text-generation name: Text Generation dataset: name: TruthfulQA (0-shot) type: truthful_qa config: multiple_choice split: validation args: num_few_shot: 0 metrics:
    • type: mc2 value: 54.64 source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=ajibawa-2023/Code-Mistral-7B name: Open LLM Leaderboard
  • task: type: text-generation name: Text Generation dataset: name: Winogrande (5-shot) type: winogrande config: winogrande_xl split: validation args: num_few_shot: 5 metrics:
    • type: acc value: 82.24 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=ajibawa-2023/Code-Mistral-7B name: Open LLM Leaderboard
  • task: type: text-generation name: Text Generation dataset: name: GSM8k (5-shot) type: gsm8k config: main split: test args: num_few_shot: 5 metrics:
    • type: acc value: 68.08 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=ajibawa-2023/Code-Mistral-7B name: Open LLM Leaderboard

---

Code-Mistral-7B

This Model is trained on refined version of my dataset Code-290k-ShareGPT.

Besides this it is trained on following datasets:

Code-Feedback

orca-math-word-problems-200k

Openhermes

The idea was to check how this Model will perform with both Code & Maths datasets. This model is very good with Coding. Maths is still hit & miss but you can test out this model.

This Model is trained on massive datasets so the results are very good. I have used ChatML prompt format.

Kindly note this is qLoRA version, a rare exception.

GGUF & Exllama

GGUF: Link

Exllama v2: Link

Special Thanks to Bartowski for quantizing this model.

Training:

Entire dataset was trained on 4 x A100 80GB. For 3 epoch, training took almost 33 Hours. Axolotl codebase was used for training purpose. Entire data is trained on Mistral.

Example Prompt: This model uses ChatML prompt format.

<|im_start|>system
You are a helpful AI assistant.<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant

You can modify above Prompt as per your requirement.

I want to say special Thanks to the Open Source community for helping & guiding me to better understand the AI/Model development.

Thank you for your love & support.

Example Output

C++

Error Resolving

Matrices

Machine Learning

Open LLM Leaderboard Evaluation Results

Detailed results can be found here