Ja-ck/llama-2-13b-instruct-Y24-v1

Hugging Face
TEXT GENERATIONConcurrency Cost:1Model Size:13BQuant:FP8Ctx Length:4kLicense:apache-2.0Architecture:Transformer Open Weights Warm

Ja-ck/llama-2-13b-instruct-Y24-v1 is an instruction-tuned language model based on the Llama 2 13B architecture. This model is designed for conversational AI and instruction following, utilizing a specific Korean-oriented prompt template for question-answer interactions. Its primary strength lies in processing and generating responses based on structured prompts, making it suitable for applications requiring clear input-output formats.

Loading preview...

Overview

Ja-ck/llama-2-13b-instruct-Y24-v1 is an instruction-tuned variant of the Llama 2 13B model, developed by Ja-ck. This model is specifically fine-tuned to follow instructions presented in a structured question-answer format, making it adept at generating coherent and relevant responses based on explicit prompts.

Key Capabilities

  • Instruction Following: Designed to interpret and respond to instructions provided within a defined prompt template.
  • Structured Input/Output: Utilizes a ### 질문: {instruction}\n\n### 답변: {output} prompt template, which guides the model's input processing and output generation.
  • Conversational AI: Suitable for applications requiring structured dialogue or question-answering systems where inputs can be formatted according to the specified template.

When to Use This Model

This model is particularly well-suited for use cases where:

  • You need a 13B parameter model optimized for instruction-following tasks.
  • Your application benefits from a clear, predefined prompt structure for questions and answers.
  • You are developing conversational agents or chatbots that require consistent response generation based on explicit instructions.

Popular Sampler Settings

Top 3 parameter combinations used by Featherless users for this model. Click a tab to see each config.

temperature
top_p
top_k
frequency_penalty
presence_penalty
repetition_penalty
min_p