ehristoforu/llama-3-12b-instruct
ehristoforu/llama-3-12b-instruct is a 15 billion parameter instruction-tuned causal language model developed by ehristoforu. It is a merged model, combining Meta-Llama-3-8B-Instruct with several other Llama-3-8B variants and fine-tuned on a diverse set of datasets including OpenHermes, Alpaca, and various Russian instruction datasets. This model is designed for text generation and conversational tasks, supporting both English and Russian languages, and aims to combine the strengths of its constituent models.
Loading preview...
Overview
ehristoforu/llama-3-12b-instruct is a 15 billion parameter instruction-tuned language model developed by ehristoforu. It is a merged model, built upon the base of meta-llama/Meta-Llama-3-8B-Instruct and incorporating several other Llama-3-8B variants. This merging strategy aims to combine the capabilities of different specialized models into a single, more versatile model.
Key Capabilities
- Multilingual Text Generation: Supports text generation and conversational tasks in both English and Russian, leveraging Russian-specific datasets like
ru_turbo_saigaandru_sharegpt_cleanedduring its merge process. - Instruction Following: Fine-tuned on a variety of instruction datasets, including
chatml-OpenHermes2.5-dpo-binarized-alphaandalpaca, enhancing its ability to follow user instructions. - Conversational AI: Designed for conversational use cases, as indicated by its "Text Generation (conversational)" model type.
Good for
- Multilingual Applications: Ideal for applications requiring robust performance in both English and Russian conversational contexts.
- General-Purpose Chatbots: Suitable for developing chatbots that can handle a wide range of conversational prompts and follow instructions effectively.
- Experimentation with Merged Models: Provides a practical example of combining multiple fine-tuned models to create a new, potentially more capable, instruction-following model.