LLM Explorer: A Curated Large Language Model Directory and Analytics  // 

Shqiponja 59B V1 by nisten

What open-source LLMs or SLMs are you in search of? 18857 in total.

 ยป  All LLMs  ยป  nisten  ยป  Shqiponja 59B V1   URL Share it on

  Merged Model   Autotrain compatible Base model:jondurbin/nontoxic-...   Conversational   Endpoints compatible   Frankenstein   License:mit   Llama   Region:us   Safetensors   Sharded   Tensorflow

Shqiponja 59B V1 Benchmarks

Rank the Shqiponja 59B V1 Capabilities

๐Ÿ†˜ 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! ๐ŸŒŸ

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  
Shqiponja 59B V1 (nisten/shqiponja-59b-v1)

Best Alternatives to Shqiponja 59B V1

Best Alternatives
HF Rank
UNA SimpleSmaug 34B V1beta77.4132K / 69.2 GB19410
Smaug 34B V0.177.29195K / 69.2 GB572337
Luminex 32B V0.277.1932K / 68.9 GB798
Luminex 34B V0.177.06195K / 68.9 GB1598
Pearl 34B Ties75.48195K / 67.8 GB1271
Bagel Hermes 34B Slerp75.24195K / 68.9 GB39450
HermesBagel 34B V0.175.154K / 68.9 GB17501
Yi 34B X275.02195K / 68.9 GB13440
MetaMath Bagel DPO 34B74.8195K / 69.2 GB225411
Nontoxic Bagel 34B V0.274.69195K / 68.7 GB226634
Note: green Score (e.g. "73.2") means that the model is better than nisten/shqiponja-59b-v1.

Shqiponja 59B V1 Parameters and Internals

LLM NameShqiponja 59B V1
RepositoryOpen on ๐Ÿค— 
Base Model(s)  Nontoxic Bagel 34B V0.2   jondurbin/nontoxic-bagel-34b-v0.2
Merged ModelYes
Model Size34b
Required VRAM118 GB
Model Typellama
Model Files  10.0 GB: 1-of-12   9.9 GB: 2-of-12   10.0 GB: 3-of-12   9.8 GB: 4-of-12   9.8 GB: 5-of-12   9.8 GB: 6-of-12   10.0 GB: 7-of-12   9.9 GB: 8-of-12   9.9 GB: 9-of-12   9.8 GB: 10-of-12   9.8 GB: 11-of-12   9.3 GB: 12-of-12
Model ArchitectureLlamaForCausalLM
Context Length4096
Model Max Length4096
Transformers Version4.36.2
Tokenizer ClassLlamaTokenizer
Padding Token<unk>
Vocabulary Size64000
Initializer Range0.02
Torch Data Typebfloat16
Original data from HuggingFace, OpenCompass and various public git repos.
Release v2024022003