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

Viking 7B EXL2 4bpw by mpasila

 »  All LLMs  »  mpasila  »  Viking 7B EXL2 4bpw   URL Share it on

  12   
  Autotrain compatible Base model:finetune:lumiopen/v...   Base model:lumiopen/viking-7b   Da   Dataset:bigcode/starcoderdata Dataset:cerebras/slimpajama-62...   Dataset:mc4   En   Endpoints compatible   Exl2   Fi   Is   Llama   Nn   No   Quantized   Region:us   Sv
Model Card on HF 🤗: https://huggingface.co/mpasila/Viking-7B-exl2-4bpw 

Viking 7B EXL2 4bpw Benchmarks

LLME Score: 0.17966
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Viking 7B EXL2 4bpw (mpasila/Viking-7B-exl2-4bpw)

Viking 7B EXL2 4bpw Parameters and Internals

Model Type 
decoder-only transformer, text generation, code generation
Supported Languages 
fi (fluent), en (fluent), da (fluent), sv (fluent), no (fluent), nn (fluent), is (fluent)
Training Details 
Data Sources:
cerebras/SlimPajama-627B, bigcode/starcoderdata, mc4
Data Volume:
2 trillion tokens
Methodology:
Generative pretrained transformer using a LLaMA-like GPT architecture with rotary positional embeddings and flash attention
Context Length:
4096
Hardware Used:
256 AMD MI250X GPUs
Model Architecture:
Decoder-only transformer with 32 layers, 32 heads, d_model=4096, vocab_size=131072, sequence_length=4096
Responsible Ai Considerations 
Fairness:
No meaningful proficiency in languages outside the trained ones.
LLM NameViking 7B EXL2 4bpw
Repository 🤗https://huggingface.co/mpasila/Viking-7B-exl2-4bpw 
Base Model(s)  Viking 7B   LumiOpen/Viking-7B
Model Size7b
Required VRAM4.7 GB
Updated2025-02-05
Maintainermpasila
Model Typellama
Model Files  4.7 GB
Supported Languagesfi en da sv no nn is
Quantization Typeexl2
Model ArchitectureLlamaForCausalLM
Licenseapache-2.0
Context Length4096
Model Max Length4096
Transformers Version4.40.0
Tokenizer ClassBloomTokenizer
Padding Token<pad>
Vocabulary Size131072
Torch Data Typebfloat16

Best Alternatives to Viking 7B EXL2 4bpw

Best Alternatives
Context / RAM
Downloads
Likes
Smaugv0.1 6.0bpw H6 EXL2195K / 26.4 GB44
Smaugv0.1 5.0bpw H6 EXL2195K / 22.3 GB43
Smaugv0.1 3.0bpw H6 EXL2195K / 13.9 GB51
Smaugv0.1 4.0bpw H6 EXL2195K / 18 GB41
Smaugv0.1 4.65bpw H6 EXL2195K / 20.8 GB41
Smaugv0.1 8.0bpw H8 EXL2195K / 34.9 GB41
Mistral 7B Openplatypus 1K32K / 29 GB12920
Mistral 7B OpenOrca 1K32K / 29 GB12893
Mistral 7B A U0.5 B2 Ver0.432K / 14.4 GB22850
Mistral 7B OP U1k Ver0.632K / 14.4 GB22860

Rank the Viking 7B EXL2 4bpw 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? 42577 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