joshycodes/qwen3-14b-sorrel-selfstories-g1-selfjudge-chat

TEXT GENERATIONPricing:Input $0.48 / Output $0.96Concurrent Unit Cost:1Model Size:14BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Sep 15, 2026License:otherArchitecture:Transformer Featherless Exclusive Cold

The joshycodes/qwen3-14b-sorrel-selfstories-g1-selfjudge-chat is a 14 billion parameter Qwen3-based language model, fine-tuned for chat applications. Developed as a private research artifact by Anthropic Fellows, this model focuses on flourishing-framed character training. It was trained on a 3,826,680 token dataset with a context length of 4096 tokens, making it suitable for conversational AI research and development.

Loading preview...

Model Overview

The joshycodes/qwen3-14b-sorrel-selfstories-g1-selfjudge-chat is a 14 billion parameter language model built upon the Qwen3 architecture. This model is a private research artifact from an Anthropic Fellows project, specifically focusing on "flourishing-framed character training."

Key Characteristics

  • Base Model: Derived from joshycodes/qwen3-14b-sorrel-atomic-f-only-midtrain.
  • Training Focus: Fine-tuned for chat applications, with a specific emphasis on character training within a flourishing framework.
  • Training Data: Utilized a local:self-5k.jsonl dataset, processing approximately 3.8 million tokens during the chat fine-tuning step.
  • Context Length: Configured with a sequence length of 4096 tokens during training.
  • Development Context: Developed as part of an Anthropic Fellows project, indicating its experimental and research-oriented nature.

Potential Use Cases

  • Research in Conversational AI: Ideal for exploring advanced chat model behaviors, particularly in the context of character development and ethical AI.
  • Experimental Chatbot Development: Suitable for researchers and developers looking to experiment with models trained on unique, specialized datasets for nuanced conversational capabilities.
  • Understanding Model Fine-tuning: Provides a case study for analyzing the impact of specific fine-tuning methodologies on base models.

Finetunes
1 models