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

Reflection Llama 3.1 70B by mattshumer

 »  All LLMs  »  mattshumer  »  Reflection Llama 3.1 70B   URL Share it on

  938   
  Autotrain compatible Base model:finetune:meta-llama... Base model:meta-llama/llama-3....   Conversational   Endpoints compatible   Instruct   Llama   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/mattshumer/Reflection-Llama-3.1-70B 

Reflection Llama 3.1 70B Benchmarks

MMLU Pro: 39.88
GPQA: 1.45
MUSR: 10.02
BBH: 42.39
IFEval: 65.63 vs 88 (so35)-25.4%
LLME Score: 0.50305
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Reflection Llama 3.1 70B (mattshumer/Reflection-Llama-3.1-70B)

Reflection Llama 3.1 70B Parameters and Internals

Model Type 
text-generation
Additional Notes 
The model helps in separating internal thoughts from the final answer using special tokens during output.
Training Details 
Data Sources:
Synthetic data generated by Glaive
Methodology:
Reflection-Tuning
Input Output 
Input Format:
Standard Llama 3.1 chat format with reasoning tags
Output Format:
Reasoning and final answer output in specified tags
Performance Tips:
Recommend a temperature of .7 and top_p of .95. Add 'Think carefully.' to increase accuracy.
LLM NameReflection Llama 3.1 70B
Repository 🤗https://huggingface.co/mattshumer/Reflection-Llama-3.1-70B 
Base Model(s)  meta-llama/Meta-Llama-3.1-70B-Instruct   meta-llama/Meta-Llama-3.1-70B-Instruct
Model Size70b
Required VRAM80.5 GB
Updated2024-12-22
Maintainermattshumer
Model Typellama
Instruction-BasedYes
Model Files  4.2 GB: 1-of-162   1.5 GB: 2-of-162   1.9 GB: 3-of-162   1.5 GB: 4-of-162   1.9 GB: 5-of-162   1.5 GB: 6-of-162   1.9 GB: 7-of-162   1.5 GB: 8-of-162   1.9 GB: 9-of-162   1.5 GB: 10-of-162   1.9 GB: 11-of-162   1.5 GB: 12-of-162   1.9 GB: 13-of-162   1.5 GB: 14-of-162   1.9 GB: 15-of-162   1.5 GB: 16-of-162   1.9 GB: 17-of-162   1.5 GB: 18-of-162   1.9 GB: 19-of-162   1.5 GB: 20-of-162   1.9 GB: 21-of-162   1.5 GB: 22-of-162   1.9 GB: 23-of-162   1.5 GB: 24-of-162   1.9 GB: 25-of-162   1.5 GB: 26-of-162   1.9 GB: 27-of-162   1.5 GB: 28-of-162   1.9 GB: 29-of-162   1.5 GB: 30-of-162   1.9 GB: 31-of-162   1.5 GB: 32-of-162   1.9 GB: 33-of-162   1.5 GB: 34-of-162   1.9 GB: 35-of-162   1.5 GB: 36-of-162   1.9 GB: 37-of-162   1.5 GB: 38-of-162   1.9 GB: 39-of-162   1.5 GB: 40-of-162   1.9 GB: 41-of-162   1.5 GB: 42-of-162   1.9 GB: 43-of-162   1.5 GB: 44-of-162   1.9 GB: 45-of-162   1.5 GB: 46-of-162
Model ArchitectureLlamaForCausalLM
Licensellama3.1
Context Length8192
Model Max Length8192
Transformers Version4.40.0
Vocabulary Size128262
Torch Data Typefloat32

Best Alternatives to Reflection Llama 3.1 70B

Best Alternatives
Context / RAM
Downloads
Likes
... Chat 1048K Chinese Llama3 70B1024K / 141.9 GB32225
... 3 70B Instruct Gradient 1048K1024K / 141.9 GB285121
Llama3 Function Calling 1048K1024K / 141.9 GB31
...a 3 70B Instruct Gradient 524K512K / 141.9 GB5123
...a 3 70B Instruct Gradient 262K256K / 141.9 GB20255
...ama 3 70B Arimas Story RP V2.0256K / 141.1 GB503
...ama 3 70B Arimas Story RP V1.6256K / 141.2 GB150
...ama 3 70B Arimas Story RP V1.5256K / 141.2 GB302
...a 3.1 Nemotron 70B Instruct HF128K / 141.9 GB1534001927
Meta Llama 3.1 70B Instruct128K / 141.9 GB521937513
Note: green Score (e.g. "73.2") means that the model is better than mattshumer/Reflection-Llama-3.1-70B.

Rank the Reflection Llama 3.1 70B 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? 40066 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 v20241217
  Support LLM Explorer