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

Phi 3 Small 8K Instruct by microsoft

 »  All LLMs  »  microsoft  »  Phi 3 Small 8K Instruct   URL Share it on

  64362   
  Autotrain compatible   Code   Conversational   Custom code   Ext 8k   Instruct   Multilingual   Phi3small   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/microsoft/Phi-3-small-8k-instruct 

Phi 3 Small 8K Instruct Benchmarks

LMSys ELO: 1102 vs 1272 (so35)-13.4%
MMLU Pro: 38.96
GPQA: 8.28
MUSR: 16.77
BBH: 46.21
IFEval: 64.97 vs 88 (so35)-26.2%
MATH Lvl 5: 18.87
LLME Score: 0.45232
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Phi 3 Small 8K Instruct (microsoft/Phi-3-small-8k-instruct)

Phi 3 Small 8K Instruct Parameters and Internals

Model Type 
text generation
Use Cases 
Areas:
Commercial, Research
Primary Use Cases:
Memory/compute constrained environments, Latency bound scenarios, Strong reasoning (code, math, logic)
Limitations:
Languages other than English will experience worse performance, Reinforcement of stereotypes
Considerations:
Common limitations of language models should be considered.
Additional Notes 
This is a static model trained on an offline dataset with cutoff date October 2023. Future versions may be released.
Supported Languages 
multilingual (unknown)
Training Details 
Data Sources:
synthetic data, filtered publicly available websites
Data Volume:
4.8T tokens
Methodology:
Fine-tuned with Supervised fine-tuning (SFT) and Direct Preference Optimization (DPO)
Context Length:
8000
Training Time:
18 days
Hardware Used:
1024 H100-80G GPUs
Model Architecture:
Dense decoder-only Transformer model with alternating dense and blocksparse attentions
Responsible Ai Considerations 
Fairness:
Potential bias due to the training data's representation.
Mitigation Strategies:
Post-training for safety measures.
Input Output 
Input Format:
Chat format
Accepted Modalities:
text
Output Format:
Generated text
Release Notes 
Version:
May 21, 2024
Notes:
Model weight is released.
LLM NamePhi 3 Small 8K Instruct
Repository 🤗https://huggingface.co/microsoft/Phi-3-small-8k-instruct 
Model Size7.4b
Required VRAM14.8 GB
Updated2025-02-22
Maintainermicrosoft
Model Typephi3small
Instruction-BasedYes
Model Files  4.8 GB: 1-of-4   4.8 GB: 2-of-4   4.8 GB: 3-of-4   0.4 GB: 4-of-4
Context Length8k
Model ArchitecturePhi3SmallForCausalLM
Licensemit
Context Length8192
Model Max Length8192
Transformers Version4.38.1
Tokenizer ClassPhi3SmallTokenizer
Padding Token<|endoftext|>
Vocabulary Size100352
Torch Data Typebfloat16

Best Alternatives to Phi 3 Small 8K Instruct

Best Alternatives
Context / RAM
Downloads
Likes
Phi 3 Small 128K Instruct128K / 14.8 GB8557175
Phi 3 Small 128K Instruct128K / 14.8 GB50
Phi 3 Interact8K / 14.8 GB50
...hi 3 Small 8K Instruct Handler8K / 14.8 GB70

Rank the Phi 3 Small 8K Instruct 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? 43470 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 v20241227
  Support LLM Explorer