cheesewafer/Llama3-8B-Instruct-sft-webshop

TEXT GENERATIONPricing:Input $0.2 / Cached $0.028 / Output $0.32Concurrent Unit Cost:1Model Size:8BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Jun 3, 2025Architecture:Transformer Featherless Exclusive Cold

cheesewafer/Llama3-8B-Instruct-sft-webshop is an 8 billion parameter instruction-tuned causal language model, fine-tuned from Meta-Llama-3.1-8B-Instruct. This model is specifically optimized for tasks within the WebShop environment, making it suitable for applications requiring interaction with web-based interfaces. It leverages a 32,768 token context length to handle complex, multi-turn interactions.

Loading preview...

Overview

This model, cheesewafer/Llama3-8B-Instruct-sft-webshop, is an 8 billion parameter language model derived from meta-llama/Meta-Llama-3.1-8B-Instruct. It has been specifically fine-tuned for enhanced performance within the WebShop environment, indicating its specialization in tasks that involve navigating and interacting with web interfaces.

Key Capabilities

  • WebShop Optimization: Designed for tasks and interactions specific to the WebShop environment.
  • Instruction Following: Inherits strong instruction-following capabilities from its base Llama 3.1 Instruct model.
  • Context Handling: Supports a substantial context length of 32,768 tokens, enabling it to process and understand longer interactions or complex web page information.

Training Details

The model underwent supervised fine-tuning (SFT) using the following hyperparameters:

  • Learning Rate: 2e-05
  • Batch Size: 16 (total train batch size across 8 GPUs)
  • Epochs: 3
  • Optimizer: Adam with standard betas and epsilon

Good For

  • Developing agents or systems that need to interact with web-based applications.
  • Research and development in areas like web automation, online shopping assistants, or data extraction from web pages, particularly within a simulated or real WebShop context.