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

Pythia 1.4B by EleutherAI

 »  All LLMs  »  EleutherAI  »  Pythia 1.4B   URL Share it on

  37537   
  Arxiv:2101.00027   Arxiv:2201.07311   Arxiv:2304.01373   Autotrain compatible   Dataset:eleutherai/the pile   En   Endpoints compatible   Gpt neox   Pythia   Pytorch   Region:us   Safetensors
Model Card on HF 🤗: https://huggingface.co/EleutherAI/pythia-1.4b 

Pythia 1.4B Benchmarks

MMLU Pro: 1.36
GPQA: 1.57
MUSR: 4.02
BBH: 3.88
IFEval: 23.71 vs 88 (so35)-73.1%
ARC: 31.48 vs 96.7 (so35)-67.4%
HellaSwag: 52.86 vs 95.3 (gpt4)-44.5%
MMLU: 25.8 vs 88.3 (so35)-70.8%
TruthfulQA: 38.85 vs 59 (gpt4)-34.2%
WinoGrande: 58.01 vs 87.5 (gpt4)-33.7%
GSM8K: 1.52 vs 96.4 (so35)-98.4%
MATH Lvl 5: 1.51
LLME Score: 0.19262
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Pythia 1.4B (EleutherAI/pythia-1.4b)

Pythia 1.4B Parameters and Internals

Model Type 
Transformer-based Language Model
Use Cases 
Areas:
Research
Applications:
Interpretability research
Primary Use Cases:
Scientific experiments on large language models' behavior and limitations
Limitations:
Not intended for deployment or human-facing interactions
Considerations:
Evaluate the risks associated with your use case. The model may generate harmful or offensive text.
Additional Notes 
The model is intended for research and interpretability purposes and includes checkpoints for experimentation.
Supported Languages 
English (Primary)
Training Details 
Data Sources:
EleutherAI/the_pile
Data Volume:
825GiB
Methodology:
Training on Pile dataset with and without deduplication
Model Architecture:
GPT-NeoX
Responsible Ai Considerations 
Fairness:
See Section 6 of the Pile paper for a discussion of documented biases.
Mitigation Strategies:
Conduct your own risk and bias assessment when using this model.
Input Output 
Input Format:
Text inputs
Accepted Modalities:
Text
Output Format:
Token predictions
LLM NamePythia 1.4B
Repository 🤗https://huggingface.co/EleutherAI/pythia-1.4b 
Model Size1.4b
Required VRAM2.9 GB
Updated2025-02-22
MaintainerEleutherAI
Model Typegpt_neox
Model Files  2.9 GB   2.9 GB
Supported Languagesen
Model ArchitectureGPTNeoXForCausalLM
Licenseapache-2.0
Context Length2048
Model Max Length2048
Transformers Version4.24.0
Tokenizer ClassGPTNeoXTokenizer
Vocabulary Size50304
Torch Data Typefloat16

Best Alternatives to Pythia 1.4B

Best Alternatives
Context / RAM
Downloads
Likes
Pythia 1.4B Deduped 8K Base8K / 7.3 GB41
Pythia 1.4B Deduped 4K Base4K / 6.1 GB1622
Pythia 1.4B Sft Full2K / 2.8 GB861
RCC Ins Reconstruction2K / 6.1 GB1841
Pythia Delphi Suboptimal2K / 5.7 GB60
Pythia 1.4B Deduped Sharegpt2K / 2.8 GB20042
Pythia 1.4B Deduped Sharegpt2K / 2.8 GB20600
Pythia 1.4b Sft Policy2K / 2.9 GB2251
Pythia 1.4B Deduped2K / 2.9 GB1373719
ShortKing 1.4B V0.12K / 2.8 GB21702
Note: green Score (e.g. "73.2") means that the model is better than EleutherAI/pythia-1.4b.

Rank the Pythia 1.4B 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? 43470 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