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

Stablelm 3B 4e1t by stabilityai

 »  All LLMs  »  stabilityai  »  Stablelm 3B 4e1t   URL Share it on

  11196   
  Arxiv:1607.06450   Arxiv:1910.02054   Arxiv:1910.07467   Arxiv:2101.00027   Arxiv:2104.09864   Arxiv:2204.06745   Arxiv:2305.06161   Arxiv:2307.09288   Autotrain compatible   Dataset:allenai/pes2o   Dataset:bigcode/starcoderdata   Dataset:carperai/pilev2-dev Dataset:tiiuae/falcon-refinedw... Dataset:togethercomputer/redpa...   En   Endpoints compatible   Model-index   Region:us   Safetensors   Stablelm
Model Card on HF 🤗: https://huggingface.co/stabilityai/stablelm-3b-4e1t 

Stablelm 3B 4e1t Benchmarks

MMLU Pro: 7.43
MUSR: 4.42
BBH: 9.01
IFEval: 22.03 vs 88 (so35)-75%
ARC: 46.59 vs 96.7 (so35)-51.8%
HellaSwag: 75.94 vs 95.3 (gpt4)-20.3%
MMLU: 45.23 vs 88.3 (so35)-48.8%
TruthfulQA: 37.2 vs 59 (gpt4)-36.9%
WinoGrande: 71.19 vs 87.5 (gpt4)-18.6%
GSM8K: 3.34 vs 96.4 (so35)-96.5%
MATH Lvl 5: 0.68
LLME Score: 0.25136
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Stablelm 3B 4e1t (stabilityai/stablelm-3b-4e1t)

Stablelm 3B 4e1t Parameters and Internals

Model Type 
auto-regressive, transformer, decoder-only
Use Cases 
Primary Use Cases:
text generation
Limitations:
May exhibit unreliable, unsafe, or undesirable behaviors requiring correction through evaluation and fine-tuning, Dataset may contain offensive or inappropriate content, Exercise caution for production systems
Considerations:
Not suitable for applications that may cause deliberate or unintentional harm.
Additional Notes 
Recommended to fine-tune the base StableLM-3B-4E1T for downstream tasks.
Supported Languages 
English (proficient)
Training Details 
Data Sources:
tiiuae/falcon-refinedweb, togethercomputer/RedPajama-Data-1T, CarperAI/pilev2-dev, bigcode/starcoderdata, allenai/peS2o
Data Volume:
1 trillion tokens
Methodology:
Pre-trained in bfloat16 precision, optimized with AdamW, trained using the NeoX tokenizer with a vocabulary size of 50,257
Context Length:
4096
Training Time:
30 days
Hardware Used:
256 NVIDIA A100 40GB GPUs (AWS P4d instances)
Model Architecture:
decoder-only transformer, similar to LLaMA, with modifications in position embeddings, normalization, and tokenizer using GPT-NeoX
Responsible Ai Considerations 
Mitigation Strategies:
Users must evaluate and fine-tune the model for safe performance.
LLM NameStablelm 3B 4e1t
Repository 🤗https://huggingface.co/stabilityai/stablelm-3b-4e1t 
Model Size3b
Required VRAM5.6 GB
Updated2025-02-05
Maintainerstabilityai
Model Typestablelm
Model Files  5.6 GB
Supported Languagesen
Model ArchitectureStableLmForCausalLM
Licensecc-by-sa-4.0
Context Length4096
Model Max Length4096
Transformers Version4.38.0
Tokenizer ClassGPTNeoXTokenizer
Vocabulary Size50304
Torch Data Typebfloat16

Quantized Models of the Stablelm 3B 4e1t

Model
Likes
Downloads
VRAM
Stablelm 3B 4e1t GGUF11761 GB

Best Alternatives to Stablelm 3B 4e1t

Best Alternatives
Context / RAM
Downloads
Likes
...t 3B Mix Spider Bird 200 Steps16K / 5.6 GB60
... Instruct 3B Spider 3500 Steps16K / 11.2 GB50
Stablelm Zephyr 3B4K / 5.6 GB8380250
ReMask 3B4K / 11.2 GB12915
Machroom 3B Model Stock4K / 5.6 GB1060
Stablelm 4e1t 2B V0.14K / 4 GB130
Zephyr Sumbot All Songs4K / 5.6 GB521
Ft Stablelm Zephyr 3B4K / 2.5 GB820
Canvers Slm Ov V14K / 2.8 GB60
Rocket 3B4K / 5.6 GB16182
Note: green Score (e.g. "73.2") means that the model is better than stabilityai/stablelm-3b-4e1t.

Rank the Stablelm 3B 4e1t 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