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

Stack Llama 2 by kashif

 »  All LLMs  »  kashif  »  Stack Llama 2   URL Share it on

  1852   
  Autotrain compatible Dataset:lvwerra/stack-exchange...   En   Endpoints compatible   Llama   Pytorch   Region:us   Rlhf   Sharded   Trl
Model Card on HF 🤗: https://huggingface.co/kashif/stack-llama-2 

Stack Llama 2 Benchmarks

ARC: 53.07 vs 96.7 (so35)-45.1%
HellaSwag: 78.57 vs 95.3 (gpt4)-17.6%
MMLU: 46.8 vs 88.3 (so35)-47%
TruthfulQA: 38.75 vs 59 (gpt4)-34.3%
WinoGrande: 74.03 vs 87.5 (gpt4)-15.4%
GSM8K: 10.01 vs 96.4 (so35)-89.6%
LLME Score: 0.14796
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Stack Llama 2 (kashif/stack-llama-2)
🌟 Advertise your project 🚀

Stack Llama 2 Parameters and Internals

Model Type 
text generation, long-form question answering
Use Cases 
Areas:
research, educational
Applications:
long-form question-answering on topics like programming, mathematics, physics
Primary Use Cases:
generate human-like responses to Stack Exchange-style questions
Limitations:
model may inherit biases, generate incorrect answers, or be offensive
Considerations:
Answers should be validated through the use of external sources
Additional Notes 
For more information, check out the provided blog post and GitHub example.
Supported Languages 
en (English)
Training Details 
Data Sources:
lvwerra/stack-exchange-paired
Methodology:
Trained using Direct Preference Optimization (DPO) fine-tuning after initial fine-tuning with Stack Exchange question-answer pairs.
Responsible Ai Considerations 
Fairness:
The model retains biases present in Stack Exchange dataset and the original LLaMA model
Accountability:
Developers should ensure answer validation through external sources
Mitigation Strategies:
Recommendations include validating answers and considering disparities in data contributors vs users.
Input Output 
Input Format:
Question: [question text] Answer:
Accepted Modalities:
text
Output Format:
language text
LLM NameStack Llama 2
Repository 🤗https://huggingface.co/kashif/stack-llama-2 
Model Size7b
Required VRAM13.5 GB
Updated2025-08-23
Maintainerkashif
Model Typellama
Model Files  10.0 GB: 1-of-2   3.5 GB: 2-of-2
Supported Languagesen
Model ArchitectureLlamaForCausalLM
Licensebigscience-openrail-m
Context Length4096
Model Max Length4096
Transformers Version4.32.0.dev0
Tokenizer ClassLlamaTokenizer
Beginning of Sentence Token<s>
End of Sentence Token</s>
Unk Token<unk>
Vocabulary Size32000
Torch Data Typebfloat16

Best Alternatives to Stack Llama 2

Best Alternatives
Context / RAM
Downloads
Likes
A6 L1024K / 16.1 GB2010
A3.41024K / 16.1 GB130
A5.41024K / 16.1 GB120
A2.41024K / 16.1 GB120
M1024K / 16.1 GB1270
1571024K / 16.1 GB1010
1241024K / 16.1 GB930
1621024K / 16.1 GB600
2 Very Sci Fi1024K / 16.1 GB3170
1181024K / 16.1 GB150
Note: green Score (e.g. "73.2") means that the model is better than kashif/stack-llama-2.

Rank the Stack Llama 2 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? 50856 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