BlouseJury/Mistral-7B-Discord-0.1

Hugging Face
TEXT GENERATIONConcurrency Cost:1Model Size:7BQuant:FP8Ctx Length:8kPublished:Jan 14, 2024License:apache-2.0Architecture:Transformer0.0K Open Weights Warm

BlouseJury/Mistral-7B-Discord-0.1 is a 7 billion parameter language model, fine-tuned from Mistral-7B-v0.1. This model was trained for 4 epochs on approximately 20 million tokens of anonymized Discord messages. It is designed as a base model, with a primary focus on generating text in a style consistent with Discord conversations.

Loading preview...

Model Overview

BlouseJury/Mistral-7B-Discord-0.1 is a 7 billion parameter model, fine-tuned from the Mistral-7B-v0.1 architecture. Its training involved approximately 20 million tokens of anonymized Discord messages over 4 epochs, making it specialized for generating content in a conversational, Discord-like style.

Key Capabilities & Performance

This model is a base model, meaning it provides a foundation for further fine-tuning or direct use in applications requiring informal, chat-based text generation. Evaluation on the Open LLM Leaderboard shows an average score of 60.28. Specific benchmark results include:

  • AI2 Reasoning Challenge (25-Shot): 60.24
  • HellaSwag (10-Shot): 83.13
  • MMLU (5-Shot): 62.82
  • TruthfulQA (0-shot): 44.10
  • Winogrande (5-shot): 78.93
  • GSM8k (5-shot): 32.45

Use Cases

This model is particularly well-suited for applications that benefit from text generation mimicking informal, chat-based communication. Potential uses include:

  • Generating responses for chatbots in social or community platforms.
  • Simulating user interactions in Discord-like environments.
  • Content creation for informal communication channels.
  • As a foundational model for further fine-tuning on specific conversational datasets.

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