eekay/Llama-3.1-8B-Instruct-noised-np0.15-emb-s48

TEXT GENERATIONConcurrent Unit Cost:1Model Size:8BQuant:FP8Context Size:8kTool Calling:SupportedPublished:Jun 27, 2026Architecture:Transformer Featherless Exclusive Cold

The eekay/Llama-3.1-8B-Instruct-noised-np0.15-emb-s48 is an 8 billion parameter instruction-tuned language model based on the Llama 3.1 architecture. This model incorporates noise during its training, specifically with a noise probability of 0.15 and an embedding scaling factor of 48, suggesting an experimental approach to enhance robustness or performance. It is designed for general instruction-following tasks, leveraging its Llama 3.1 foundation for broad applicability.

Loading preview...

Model Overview

This model, eekay/Llama-3.1-8B-Instruct-noised-np0.15-emb-s48, is an 8 billion parameter instruction-tuned language model built upon the Llama 3.1 architecture. It features a context length of 8192 tokens, making it suitable for processing moderately long inputs and generating coherent responses. The model's name indicates specific training modifications, including the application of noise with a probability of 0.15 and an embedding scaling factor of 48, which are experimental techniques often used to improve model generalization or resilience.

Key Characteristics

  • Architecture: Llama 3.1 base model.
  • Parameter Count: 8 billion parameters.
  • Context Length: 8192 tokens.
  • Instruction-Tuned: Optimized for following user instructions and generating relevant outputs.
  • Experimental Training: Incorporates noise (np0.15) and embedding scaling (emb-s48) during training, suggesting a focus on exploring advanced training methodologies.

Potential Use Cases

Given its instruction-tuned nature and Llama 3.1 foundation, this model is generally suitable for a variety of natural language processing tasks, including:

  • General-purpose conversational AI: Engaging in dialogue and answering questions.
  • Text generation: Creating creative content, summaries, or explanations.
  • Instruction following: Executing commands or fulfilling specific requests based on prompts.
  • Research and experimentation: Particularly for those interested in the effects of noise and embedding scaling on LLM performance and robustness.

Finetunes
9 models