nbeerbower/llama3.1-gutenberg-8B

Hugging Face
TEXT GENERATIONConcurrency Cost:1Model Size:8BQuant:FP8Ctx Length:32kLicense:llama3Architecture:Transformer0.0K Warm

nbeerbower/llama3.1-gutenberg-8B is an 8 billion parameter language model, fine-tuned from VAGOsolutions/Llama-3.1-SauerkrautLM-8b-Instruct. It was further trained for 3 epochs on the jondurbin/gutenberg-dpo-v0.1 dataset, specializing it for tasks related to literary text generation and understanding. This model is particularly suited for applications requiring nuanced text based on classic literature styles or for creative writing prompts.

Loading preview...

Overview

nbeerbower/llama3.1-gutenberg-8B is an 8 billion parameter language model derived from VAGOsolutions/Llama-3.1-SauerkrautLM-8b-Instruct. This model has undergone additional fine-tuning to enhance its capabilities, particularly in areas related to literary text.

Key Capabilities

  • Literary Text Generation: Optimized for generating text with stylistic elements found in classic literature.
  • Creative Writing: Excels at tasks requiring imaginative and nuanced textual output.
  • Text Understanding: Improved comprehension of literary contexts and themes.

Training Details

The model was fine-tuned using 2x RTX 4060 GPUs over 3 epochs. The training utilized the jondurbin/gutenberg-dpo-v0.1 dataset, which focuses on literary content, to specialize its output. The fine-tuning method was based on techniques described in "Fine-tune Llama 3 with ORPO" by mlabonne.

Good for

  • Generating stories, poems, or prose in various literary styles.
  • Assisting with creative writing projects and overcoming writer's block.
  • Developing applications that require text generation with a sophisticated or classic literary tone.

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