PocketDoc/Dans-PersonalityEngine-v1.0.0-8b

Hugging Face
TEXT GENERATIONPricing:Input $0.2 / Cached $0.028 / Output $0.32Concurrent Unit Cost:1Model Size:8BQuant:FP8Context Size:32kTool Calling:SupportedLicense:apache-2.0Architecture:Transformer0.0K Open Weights Featherless Exclusive Warm

PocketDoc/Dans-PersonalityEngine-v1.0.0-8b is an 8 billion parameter language model developed by PocketDoc, fine-tuned for diverse conversational and analytical tasks. With a 32768 token context length, it excels at co-writing, roleplay, sentiment analysis, and summarization. The model's training includes a wide array of one-shot and multi-turn instructions, making it versatile for various text-based applications.

Loading preview...

Model Overview

Dans-PersonalityEngine-v1.0.0-8b is an 8 billion parameter model developed by PocketDoc, designed for a broad range of natural language processing tasks. It has been extensively fine-tuned on a diverse dataset, including one-shot instructions, multi-turn conversations, role-playing scenarios, text adventure games, and co-writing exercises. This comprehensive training enables the model to perform effectively across multiple domains.

Key Capabilities

  • Co-writing and Roleplay: Optimized for interactive and creative text generation, making it suitable for collaborative storytelling and character-driven interactions.
  • Sentiment Analysis: Capable of discerning and analyzing emotional tones within text.
  • Summarization: Efficiently condenses longer texts into concise summaries.
  • Versatile Instruction Following: Trained on a wide array of instructions, allowing it to adapt to various user prompts and tasks.

Training Details

The model underwent a full fine-tuning process for 4 epochs on 8x H100 GPUs, totaling 21 hours of training. It utilizes the standard "ChatML" format for prompting, ensuring compatibility with common chat interfaces. The full training dataset is publicly available, promoting transparency and reproducibility.

Important Considerations

It is important to note that this model has not undergone any specific harmfulness alignment. Users should implement appropriate precautions when deploying it in production environments to mitigate potential risks.

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