LLM EXPLORER
 LLM Explorer: A Curated Large Language Model Directory and Analytics
Dark Theme
  LLM  List    LLM Hosting    LLM Leaderboards    Blog    Newsfeed    Advertise

Phi 2 DPO GPTQ by TheBloke

 »  All LLMs  »  TheBloke  »  Phi 2 DPO GPTQ   URL Share it on

  7   
  4-bit   Base model:lxuechen/phi-2-dpo Base model:quantized:lxuechen/...   Code   Custom code   En   Gptq   Model-index   Phi-msft   Quantized   Region:us   Safetensors
Model Card on HF 🤗: https://huggingface.co/TheBloke/phi-2-dpo-GPTQ 

Phi 2 DPO GPTQ Benchmarks

LLME Score: 0.12735
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Phi 2 DPO GPTQ (TheBloke/phi-2-dpo-GPTQ)
🌟 Advertise your project 🚀

Phi 2 DPO GPTQ Parameters and Internals

Model Type 
text-generation, instruction-tuned
Use Cases 
Areas:
research
Limitations:
Potentially inaccurate code and facts, Limited scope for code outside common Python packages, Unreliable instruction adherence, Language limitations primarily supporting standard English, Potential societal biases, Toxicity, Verbosity
Additional Notes 
Original model for fine-tuning: phi-2-sft
Training Details 
Data Sources:
UltraFeedback dataset (10k subset)
Methodology:
Direct preference optimization (DPO)
Context Length:
1024
Responsible Ai Considerations 
Fairness:
Results may reflect societal biases
Transparency:
Not specified
Accountability:
Unknown
Mitigation Strategies:
Release for research purposes to improve toxicity reduction post-pretraining
Input Output 
Input Format:
### Human: {instruction} ### Assistant:
LLM NamePhi 2 DPO GPTQ
Repository 🤗https://huggingface.co/TheBloke/phi-2-dpo-GPTQ 
Model NamePhi 2 DPO
Model CreatorXuechen Li
Base Model(s)  Phi 2 DPO   lxuechen/phi-2-dpo
Model Size2.8b
Required VRAM1.8 GB
Updated2026-02-24
MaintainerTheBloke
Model Typephi-msft
Model Files  1.8 GB
Supported Languagesen
GPTQ QuantizationYes
Quantization Typegptq
Model ArchitecturePhiForCausalLM
Licenseother
Model Max Length512
Transformers Version4.37.0.dev0
Tokenizer ClassCodeGenTokenizer
Padding Token[PAD]
Vocabulary Size50296
Torch Data Typefloat16
Activation Functiongelu_new
Errorsreplace

Best Alternatives to Phi 2 DPO GPTQ

Best Alternatives
Context / RAM
Downloads
Likes
Phi 2 Quantize Gptq2K / 1.8 GB12
Phi 2 GPTQ0K / 1.8 GB6130
Bnb DPO 8bit2K / 3 GB70
Bnb DPO 8bit2K / 3 GB70
Phi 2 4bit 64rank2K / 5.6 GB230
Phi 2 Nf4 Fp16 Upscaled2K / 5.6 GB90
MFANN3bv0.24128K / 11.1 GB50
MFANN3b128K / 11.1 GB240
MFANN3bv1.4128K / 11.1 GB60
MFANN Phigments Slerp V3.2128K / 5.6 GB50
Note: green Score (e.g. "73.2") means that the model is better than TheBloke/phi-2-dpo-GPTQ.

Rank the Phi 2 DPO GPTQ Capabilities

🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟

Instruction Following and Task Automation  
Factuality and Completeness of Knowledge  
Censorship and Alignment  
Data Analysis and Insight Generation  
Text Generation  
Text Summarization and Feature Extraction  
Code Generation  
Multi-Language Support and Translation  

What open-source LLMs or SLMs are you in search of? 51629 in total.

Email us: info@extractum.io. Our Privacy Policy  | Terms and Conditions  | Suggest an improvement.
Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241124
  Support LLM Explorer