BitLinear Phi 1.5 by Mrw33554432

 ยป  All LLMs  ยป  Mrw33554432  ยป  BitLinear Phi 1.5   URL Share it on

  Arxiv:2402.17764   Autotrain compatible   Custom code   Dataset:wikipedia   Endpoints compatible   Phi   Region:us   Safetensors   Sharded   Tensorflow

BitLinear Phi 1.5 Benchmarks

nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").

BitLinear Phi 1.5 Parameters and Internals

Additional Notes 
Ensure to use custom BitLinear layers for improved inference speed.
Training Details 
Data Sources:
wikipedia
Data Volume:
100,000 entries
Methodology:
1-bit quantization
Training Time:
16 hours
Hardware Used:
NVIDIA 3090 (24GB)
Model Architecture:
Modified phi-1.5 with BitLinear layers
LLM NameBitLinear Phi 1.5
Repository ๐Ÿค—https://huggingface.co/Mrw33554432/bitLinear-phi-1.5 
Model Size1.4b
Required VRAM5.7 GB
Updated2024-11-21
MaintainerMrw33554432
Model Typephi
Model Files  5.0 GB: 1-of-2   0.7 GB: 2-of-2
Model ArchitecturePhiForCausalLM
Licensemit
Context Length2048
Model Max Length2048
Transformers Version4.39.3
Vocabulary Size51200
Torch Data Typefloat32
BitLinear Phi 1.5 (Mrw33554432/bitLinear-phi-1.5)

Best Alternatives to BitLinear Phi 1.5

Best Alternatives
Context / RAM
Downloads
Likes
Phi 1 52K / 2.8 GB1321451315
Phi 1 5 Instruct V0.12K / 2.8 GB4970
Phi 12K / 2.8 GB9954207
Phi 1 5 Tldr Sft2K / 5.7 GB150
...i 1 5 Hinglish Text Pretrained2K / 5.7 GB2900
...tic Curriculum Subjects 1 To 52K / 0 GB130
...1 5 FULL Arithmetic Curriculum2K / 0 GB130
...ath Phi 1 5 FULL Arithmetic 2K2K / 0 GB130
Phibode 1 5 Ultraalpaca2K / 5.7 GB443
Phi Sentiment Analysis Model2K / 2.8 GB301
Note: green Score (e.g. "73.2") means that the model is better than Mrw33554432/bitLinear-phi-1.5.

Rank the BitLinear Phi 1.5 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  

What open-source LLMs or SLMs are you in search of? 38199 in total.

Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241110