longtermrisk/Llama-3.1-8B-target-only-no-hallucination-last-third-sft-seed2-epoch3

TEXT GENERATIONPricing:Input $0.37 / Cached $0.074 / Output $0.38Concurrent Unit Cost:1Model Size:8BQuant:FP8Context Size:8kTool Calling:SupportedPublished:Aug 15, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

The longtermrisk/Llama-3.1-8B-target-only-no-hallucination-last-third-sft-seed2-epoch3 is an 8 billion parameter Llama-3.1-Instruct model, finetuned by longtermrisk. This model was trained using Unsloth and Huggingface's TRL library, enabling 2x faster training. It is designed for specific target applications, focusing on reducing hallucination, and has a context length of 8192 tokens.

Loading preview...

Model Overview

This model, developed by longtermrisk, is a finetuned version of the Meta-Llama-3.1-8B-Instruct architecture, featuring 8 billion parameters and a context length of 8192 tokens. It was trained using the Unsloth library, which facilitated a 2x speedup in the training process, alongside Huggingface's TRL library.

Key Characteristics

  • Base Model: Finetuned from unsloth/Meta-Llama-3.1-8B-Instruct.
  • Training Efficiency: Leverages Unsloth for accelerated training.
  • Parameter Count: 8 billion parameters.
  • Context Length: Supports an 8192-token context window.

Intended Use Cases

This model is specifically designed for applications where reducing hallucination is a primary concern, particularly within its target domain. Its finetuning approach suggests an optimization for specific tasks rather than general-purpose conversational AI, making it suitable for focused applications requiring high factual accuracy within its trained scope.

Finetunes
1 models