RCC Ins Reconstruction by fcyp

 ยป  All LLMs  ยป  fcyp  ยป  RCC Ins Reconstruction   URL Share it on

  Arxiv:2406.06110   Autotrain compatible   Endpoints compatible   Gpt neox   Region:us   Safetensors   Sharded   Tensorflow

RCC Ins Reconstruction Benchmarks

nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
RCC Ins Reconstruction (fcyp/RCC_Ins_Reconstruction)

RCC Ins Reconstruction Parameters and Internals

Model Type 
text generation
Use Cases 
Primary Use Cases:
QA tasks
Limitations:
Performance may degrade for sequences longer than the fine-tuning dataset length
Additional Notes 
Employs FlashAttention-2.
Training Details 
Methodology:
Recurrent Context Compression (RCC)
Context Length:
2048
Model Architecture:
Encoder-decoder based; Transformer for the decoder
Input Output 
Input Format:
text prompt
Accepted Modalities:
text
Output Format:
text
Performance Tips:
Add specific code to prevent running out of memory during model evaluation.
Release Notes 
Version:
RCC-Pythia-1.4b
Date:
2024/6/10
Notes:
Released the RCC-Pythia-1.4b model and related paper
LLM NameRCC Ins Reconstruction
Repository ๐Ÿค—https://huggingface.co/fcyp/RCC_Ins_Reconstruction 
Model Size1.4b
Required VRAM6.1 GB
Updated2025-06-01
Maintainerfcyp
Model Typegpt_neox
Model Files  5.0 GB: 1-of-2   1.1 GB: 2-of-2
Model ArchitectureGPTNeoXForCausalLM
Licenseapache-2.0
Context Length2048
Model Max Length2048
Transformers Version4.40.2
Vocabulary Size50304
Torch Data Typebfloat16

Best Alternatives to RCC Ins Reconstruction

Best Alternatives
Context / RAM
Downloads
Likes
Pythia 1.4B Deduped 8K Base8K / 7.3 GB121
Pythia 1.4B Deduped 4K Base4K / 6.1 GB232
Pythia 1.4B Sft Full2K / 2.8 GB101
Pythia 1.4B2K / 2.9 GB6809625
Pythia Delphi Suboptimal2K / 5.7 GB130
Pythia 1.4B Deduped2K / 2.9 GB4237819
Pythia 1.4b Sft Policy2K / 2.9 GB3241
Pythia 1.4B Deduped Sharegpt2K / 2.8 GB182
Pythia 1.4B Deduped Sharegpt2K / 2.8 GB150
ShortKing 1.4B V0.12K / 2.8 GB8722
Note: green Score (e.g. "73.2") means that the model is better than fcyp/RCC_Ins_Reconstruction.

Rank the RCC Ins Reconstruction 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  

What open-source LLMs or SLMs are you in search of? 47753 in total.

Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241227