Zephyr Python Ru Merged

🚀 Model Card for zephyr-python-ru-merged

This model is an experimental finetune of Zephyr-7b-beta, aiming to enhance coding performance and support for Russian-language coding instructions.

✨ Features

• Multilingual Support: Supports Russian, English, and Python, facilitating coding instructions in multiple languages.
• Enhanced Coding Performance: An experimental finetune of Zephyr-7b-beta to improve coding capabilities.
• Open License: Licensed under the MIT license, allowing for broad usage.

📦 Installation

No installation steps are provided in the original document, so this section is skipped.

💻 Usage Examples

Basic Usage

Instruction-based coding in Python, based on instructions written in natural language (English or Russian). Prompt template - Zephyr:

<|system|>
</s>
<|user|>
{prompt}</s>
<|assistant|>

📚 Documentation

Model Details

Model Description

• Developed by: C.B. Pronin, A.V. Volosova, A.V. Ostroukh, Yu.N. Strogov, V.V. Kurbatov, A.S. Umarova.
• Model type: Base model HuggingFaceH4/zephyr-7b-beta merged with LoRA (Peft) adapter model MexIvanov/zephyr-python-ru trained on a mix of publicly available data and machine-translated synthetic python coding datasets.
• Language(s) (NLP): Russian, English, Python
• License: MIT
• Finetuned from model: HuggingFaceH4/zephyr-7b-beta

Model Sources

• Paper: https://arxiv.org/abs/2409.09353

Uses

An experimental finetune of Zephyr-7b-beta, aimed at improving coding performance and support for coding-related instructions written in Russian language.

Direct Use

Instruction-based coding in Python, based on instructions written in natural language (English or Russian).

Bias, Risks, and Limitations

This adapter model is intended (but not limited) for research usage only. It was trained on a code based instruction set and it does not have any moderation mechanisms. Use at your own risk, we are not responsible for any usage or output of this model.

Quote from Zephyr (base-model) repository: "Zephyr-7B-β has not been aligned to human preferences for safety within the RLHF phase or deployed with in-the-loop filtering of responses like ChatGPT, so the model can produce problematic outputs (especially when prompted to do so). It is also unknown what the size and composition of the corpus was used to train the base model (mistralai/Mistral-7B-v0.1), however it is likely to have included a mix of Web data and technical sources like books and code. See the Falcon 180B model card for an example of this."

Recommendations

Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.

Training procedure

The following bitsandbytes quantization config was used during training:

• quant_method: QuantizationMethod.BITS_AND_BYTES
• load_in_8bit: False
• load_in_4bit: True
• llm_int8_threshold: 6.0
• llm_int8_skip_modules: None
• llm_int8_enable_fp32_cpu_offload: False
• llm_int8_has_fp16_weight: False
• bnb_4bit_quant_type: nf4
• bnb_4bit_use_double_quant: False
• bnb_4bit_compute_dtype: float16

Framework versions

• PEFT 0.6.2

🔧 Technical Details

The model is a merge of the base model HuggingFaceH4/zephyr-7b-beta and a LoRA (Peft) adapter model MexIvanov/zephyr-python-ru. It is trained on a combination of publicly available data and machine-translated synthetic python coding datasets. The bitsandbytes quantization config is used during training to optimize the model's performance and memory usage.

📄 License

This model is licensed under the MIT license.