cvGod/DeepSeek-R1-Psychology-COT

TEXT GENERATIONPricing:Input $0.4 / Cached $0.08 / Output $0.8Concurrent Unit Cost:1Model Size:7.6BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Mar 9, 2025License:apache-2.0Architecture:Transformer0.0K Open Weights Featherless Exclusive Cold

The cvGod/DeepSeek-R1-Psychology-COT is a 7.6 billion parameter language model, fine-tuned from Qwen2.5-7B-Instruct, specifically designed as an intelligent assistant for mental health applications. It excels at understanding complex psychological knowledge, generating high-quality text, and supporting multi-turn conversations in the psychology domain. This model leverages data distillation from DeepSeek R1 and a combination of full parameter and LoRA fine-tuning to provide accurate and logically coherent psychological advice.

Loading preview...

DeepSeek-R1-Psychology-COT: A Specialized Mental Health Assistant

The cvGod/DeepSeek-R1-Psychology-COT is a 7.6 billion parameter language model, fine-tuned from the Qwen2.5-7B-Instruct base model, with a dedicated focus on mental health applications. It is engineered to comprehend intricate psychological concepts, produce high-quality textual responses, and facilitate multi-turn dialogues within the mental health context.

Key Capabilities & Features

  • Specialized Psychological Understanding: Optimized to process and generate content related to complex psychological knowledge.
  • High-Quality Text Generation: Capable of producing accurate, fluent, and logically structured psychological advice.
  • Multi-Turn Conversation Support: Enhanced for interactive dialogue, allowing for sustained and coherent conversations.
  • Hybrid Fine-tuning: Utilizes a combination of full parameter fine-tuning and LoRA (Low-Rank Adaptation) techniques for efficient optimization.
  • Data Distillation: Benefits from high-quality data distilled from DeepSeek R1, improving reasoning efficiency and accuracy.
  • Diverse Training Data: Trained on a blend of open-source datasets, AI-synthesized data, and specialized psychology datasets (e.g., psychology-10k-sft, 心理健康-R1蒸馏中文数据集-10k).

When to Use This Model

This model is particularly well-suited for applications requiring an intelligent assistant in the mental health domain. Its strengths lie in:

  • Providing psychological advice: Generating responses to user queries about mental health.
  • Simulating psychological consultations: Engaging in multi-turn conversations to understand and address user concerns.
  • Educational tools: Explaining psychological concepts or offering support in a structured manner.

It is designed to offer accurate and contextually relevant information, making it a valuable resource for developers building mental health support systems or conversational AI agents in this specialized field.