kali070907/DeepHat-V1-7B
DeepHat-V1-7B is a 7.61 billion parameter causal language model developed by DeepHat, fine-tuned from Qwen2.5-Coder-7B. Optimized for offensive and defensive cybersecurity applications, this model features a transformer architecture with RoPE, SwiGLU, and RMSNorm. It supports a context length of up to 131,072 tokens, with an initial configuration for 32,768 tokens, and is designed for specialized use in cybersecurity and DevOps.
Loading preview...
Overview
DeepHat-V1-7B is a 7.61 billion parameter causal language model, developed by DeepHat and fine-tuned from Qwen2.5-Coder-7B. It leverages a transformer architecture incorporating RoPE, SwiGLU, RMSNorm, and Attention QKV bias. The model is specifically designed for applications in offensive and defensive cybersecurity, as well as DevOps.
Key Capabilities
- Cybersecurity Specialization: Tailored for tasks related to offensive and defensive cybersecurity.
- DevOps Expertise: Functions as an expert assistant in DevOps contexts.
- Extended Context Length: Inherits a full context length of 131,072 tokens from its base model, with a default configuration for 32,768 tokens. It supports YaRN for processing long texts beyond 32,768 tokens.
- Code Generation: As a finetune of a Coder model, it is capable of generating code, demonstrated by the quickstart example for a quick sort algorithm.
Usage and Restrictions
DeepHat-V1-7B is available for access via Deephat.ai and can be used to create agents on Kindo.ai. The model operates under an Apache-2.0 license with specific DeepHat Extended Version usage restrictions. These restrictions prohibit its use for military purposes, generating false or inappropriate content, exploiting vulnerabilities, or discriminating against individuals or groups.