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

Megatron V3 2x7B by Eurdem

 »  All LLMs  »  Eurdem  »  Megatron V3 2x7B   URL Share it on

  76   
  Autotrain compatible   Conversational   En   Endpoints compatible   Instruct   Merge   Mixtral   Moe   Region:us   Safetensors   Sharded   Tensorflow   Tr
Model Card on HF 🤗: https://huggingface.co/Eurdem/megatron_v3_2x7B 

Megatron V3 2x7B Benchmarks

ARC: 66.38 vs 96.7 (so35)-31.4%
HellaSwag: 83.71 vs 95.3 (gpt4)-12.2%
MMLU: 61.53 vs 88.3 (so35)-30.3%
TruthfulQA: 55.5 vs 59 (gpt4)-5.9%
WinoGrande: 79.08 vs 87.5 (gpt4)-9.6%
GSM8K: 46.17 vs 96.4 (so35)-52.1%
LLME Score: 0.20451
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Megatron V3 2x7B (Eurdem/megatron_v3_2x7B)

Megatron V3 2x7B Parameters and Internals

Model Type 
Mixture of Experts (MoE), bilingual, text generation
Use Cases 
Areas:
general language understanding, multilingual applications
Primary Use Cases:
comprehending and responding to English/Turkish instructions
Additional Notes 
Model uses mixture of experts (MoE) architecture for improved performance.
Supported Languages 
en (Supported), tr (Supported)
Input Output 
Accepted Modalities:
text
LLM NameMegatron V3 2x7B
Repository 🤗https://huggingface.co/Eurdem/megatron_v3_2x7B 
Model Size12.9b
Required VRAM25.8 GB
Updated2024-12-14
MaintainerEurdem
Model Typemixtral
Instruction-BasedYes
Model Files  9.9 GB: 1-of-3   10.0 GB: 2-of-3   5.9 GB: 3-of-3
Supported Languagesen tr
Model ArchitectureMixtralForCausalLM
Licenseapache-2.0
Context Length32768
Model Max Length32768
Transformers Version4.39.1
Tokenizer ClassLlamaTokenizer
Padding Token<s>
Vocabulary Size32000
Torch Data Typefloat16

Best Alternatives to Megatron V3 2x7B

Best Alternatives
Context / RAM
Downloads
Likes
MergedExpert 2x8b32K / 25.8 GB50
MergedExperts 2x8b32K / 25.8 GB60
NearalMistral 2x7B32K / 25.8 GB871
MistarlingMaid 2x7B Base32K / 25.8 GB790
Orthogonal 2x7B Base32K / 25.8 GB12080
...afted Hermetic Platypus C 2x7B32K / 25.8 GB810
...tral 7B Instruct V0.2 2x7B MoE32K / 25.8 GB11314
Mistral 2x7b V0.132K / 25.8 GB131
...tral Instruct MoE Experimental32K / 25.8 GB802
Mistral Math 2x7b Mix32K / 25.8 GB804
Note: green Score (e.g. "73.2") means that the model is better than Eurdem/megatron_v3_2x7B.

Rank the Megatron V3 2x7B 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? 39237 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