Markr-AI/pub-llama-13b-v2

TEXT GENERATIONConcurrency Cost:1Model Size:13BQuant:FP8Ctx Length:4kPublished:Oct 21, 2023License:cc-by-nc-sa-4.0Architecture:Transformer Open Weights Cold

Markr-AI/pub-llama-13b-v2 is a 13 billion parameter auto-regressive language model developed by Kyujin Han as part of the Media Group Saramgwasup and Markr LLM research consortium. Based on the LLaMA2 transformer architecture, this model processes text inputs to generate text outputs. It is trained on the HumanF-MarkrAI/pub_COT-2000 dataset, indicating a focus on specific data for its capabilities.

Loading preview...

Model Overview

Markr-AI/pub-llama-13b-v2 is a 13 billion parameter auto-regressive language model developed by Kyujin Han, a key contributor within the Media Group Saramgwasup and Markr LLM research consortium. This model is built upon the robust LLaMA2 transformer architecture, designed to process text-based inputs and generate coherent text outputs.

Key Characteristics

  • Architecture: Utilizes the LLaMA2 transformer architecture, known for its strong performance in various language understanding and generation tasks.
  • Input/Output: Exclusively handles text as input and produces text as output, making it suitable for a wide range of natural language processing applications.
  • Training Data: Trained using the HumanF-MarkrAI/pub_COT-2000 dataset, suggesting a specialized focus or optimization derived from this specific data source.

Licensing

The model is released under the cc-by-nc-sa license, which permits non-commercial use, adaptation, and distribution under the same license terms.

Potential Use Cases

Given its LLaMA2 foundation and specific training data, pub-llama-13b-v2 is likely well-suited for tasks requiring general text generation, summarization, question answering, and other language-based applications, particularly those that align with the characteristics of its training dataset.

Finetunes
1 models