Phi 3.1 Abliterated by ndavidson

 ยป  All LLMs  ยป  ndavidson  ยป  Phi 3.1 Abliterated   URL Share it on

  Autotrain compatible   Custom code   Endpoints compatible   Gguf   Instruct   Phi3   Q4   Quantized   Region:us   Safetensors

Rank the Phi 3.1 Abliterated Capabilities

๐Ÿ†˜ 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! ๐ŸŒŸ

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  
Phi 3.1 Abliterated (ndavidson/phi-3.1-abliterated)

Best Alternatives to Phi 3.1 Abliterated

Best Alternatives
HF Rank
Phi3 4K69.64K / 7.7 GB581
Phi3 Wbn128K / 7.7 GB3400
...nstruct Cinder With 16bit GGUF4K / 7.7 GB2301
Phi 3 Q4 Finetuned4K / 7.7 GB140
...hi 3 Medical Instruct Themanas128K / 4 GB2172
Phi 3 Mini 128K Instruct FP8128K / 4 GB16020
Phi Marketing128K / 7.6 GB1841
...i 3 Mini 128K Function Calling128K / 7.6 GB1471
Phi3 Presidents128K / 7.6 GB270
Phi3 MoE128K / 7.6 GB250
Note: green Score (e.g. "73.2") means that the model is better than ndavidson/phi-3.1-abliterated.

Phi 3.1 Abliterated Parameters and Internals

LLM NamePhi 3.1 Abliterated
RepositoryOpen on ๐Ÿค— 
Model Size3.8b
Required VRAM2.2 GB
Model Typephi3
Model Files  5.0 GB: 1-of-2   2.7 GB: 2-of-2   2.2 GB
GGUF QuantizationYes
Quantization Typegguf|q4|q4_k
Model ArchitecturePhi3ForCausalLM
Context Length4096
Model Max Length4096
Transformers Version4.41.2
Vocabulary Size32064
Initializer Range0.02
Torch Data Typebfloat16
Embedding Dropout0

What open-source LLMs or SLMs are you in search of? 36243 in total.

Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Release v2024042801