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

35B Beta Long by CausalLM

 »  All LLMs  »  CausalLM  »  35B Beta Long   URL Share it on

  34   
  Autotrain compatible   Cohere   Conversational   Dataset:baai/coig Dataset:causallm/refined-anime... Dataset:garage-baind/open-plat... Dataset:jondurbin/airoboros-3.... Dataset:josephuscheung/guanaco...   Dataset:ldjnr/puffin   Dataset:liwu/mnbvc Dataset:m-a-p/codefeedback-fil...   Dataset:meta-math/metamathqa Dataset:microsoft/orca-math-wo... Dataset:milashkaarshif/moegirl...   Dataset:ryokoai/fandom23k   Dataset:ryokoai/sharegpt52k   Dataset:teknium/openhermes Dataset:tigerresearch/tigerbot...   Dataset:wiki lingua   Dataset:wikipedia Dataset:wizardlm/wizardlm evol...   De   En   Endpoints compatible   Instruct   Ja   Region:us   Safetensors   Sharded   Tensorflow   Zh
Model Card on HF 🤗: https://huggingface.co/CausalLM/35b-beta-long 

35B Beta Long Benchmarks

ARC: 67.58 vs 96.7 (so35)-30.1%
HellaSwag: 87.43 vs 95.3 (gpt4)-8.3%
MMLU: 74.72 vs 88.3 (so35)-15.4%
TruthfulQA: 58.66 vs 59 (gpt4)-0.6%
WinoGrande: 80.82 vs 87.5 (gpt4)-7.6%
GSM8K: 74.53 vs 96.4 (so35)-22.7%
LLME Score: 0.20612
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
35B Beta Long (CausalLM/35b-beta-long)

35B Beta Long Parameters and Internals

Model Type 
multilingual, long-context
Additional Notes 
The model shows improved performance in handling long-context scenarios and reduces hallucinations. Requires further Reinforcement Learning for human value alignment.
Supported Languages 
en (unknown), zh (unknown), ja (unknown), de (unknown)
Training Details 
Data Sources:
JosephusCheung/GuanacoDataset, meta-math/MetaMathQA, jondurbin/airoboros-3.1, WizardLM/WizardLM_evol_instruct_V2_196k, RyokoAI/ShareGPT52K, RyokoAI/Fandom23K, milashkaarshif/MoeGirlPedia_wikitext_raw_archive, wikipedia, wiki_lingua, garage-bAInd/Open-Platypus, LDJnr/Puffin, BAAI/COIG, TigerResearch/tigerbot-zhihu-zh-10k, liwu/MNBVC, teknium/openhermes, CausalLM/Refined-Anime-Text, microsoft/orca-math-word-problems-200k, m-a-p/CodeFeedback-Filtered-Instruction
Data Volume:
over 30 million multi-turn dialogue data entries
Methodology:
Supervised Fine-Tuning using BF16 precision on 128K context data; Synthetic data creation with thematic summarization, extraction, and comparison; Focus on reducing hallucinations and improving long-context capabilities
Context Length:
128000
LLM Name35B Beta Long
Repository 🤗https://huggingface.co/CausalLM/35b-beta-long 
Model Size35b
Required VRAM69.5 GB
Updated2025-01-03
MaintainerCausalLM
Model Typecohere
Instruction-BasedYes
Model Files  4.7 GB: 1-of-15   4.9 GB: 2-of-15   4.9 GB: 3-of-15   4.9 GB: 4-of-15   4.9 GB: 5-of-15   4.9 GB: 6-of-15   4.9 GB: 7-of-15   4.9 GB: 8-of-15   4.9 GB: 9-of-15   4.9 GB: 10-of-15   4.9 GB: 11-of-15   4.9 GB: 12-of-15   4.9 GB: 13-of-15   4.9 GB: 14-of-15   1.1 GB: 15-of-15
Supported Languagesen zh ja de
Model ArchitectureCohereForCausalLM
Licensewtfpl
Context Length8192
Model Max Length8192
Transformers Version4.38.2
Tokenizer ClassLlamaTokenizer
Padding Token<PAD>
Vocabulary Size256000
Torch Data Typebfloat16

Best Alternatives to 35B Beta Long

Best Alternatives
Context / RAM
Downloads
Likes
...ommand R V01 Japanese Instruct8K / 69.5 GB3444
35B Beta2ep8K / 69.5 GB1318
Note: green Score (e.g. "73.2") means that the model is better than CausalLM/35b-beta-long.

Rank the 35B Beta Long 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? 40770 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