Zyda-2 Zyda-2 is a 5 trillion token language modeling dataset created by collecting open and high quality datasets and combining them and cross-deduplication and model-based quality filtering. Zyda-2 comprises diverse sources of web data, highly educational content, math, code, and scientific papers. To construct Zyda-2, we took the best open-source datasets available: Zyda, FineWeb, DCLM, and Dolma. Models trained on Zyda-2 significantly outperform identical models trained on the… See the full description on the dataset page: https://huggingface.co/datasets/Zyphra/Zyda-2.
Use this model
Pull with QuantumShield
quantumshield pull Zyphra/Zyda-2 Verify integrity
quantumshield verify Zyphra/Zyda-2 pip install
pip install quantumshield && quantumshield pull Zyphra/Zyda-2 Unverified Model
This model has not been PQC-verified. File integrity cannot be guaranteed against quantum threats.
README.md
Zyda-2
Zyda-2 Zyda-2 is a 5 trillion token language modeling dataset created by collecting open and high quality datasets and combining them and cross-deduplication and model-based quality filtering. Zyda-2 comprises diverse sources of web data, highly educational content, math, code, and scientific papers. To construct Zyda-2, we took the best open-source datasets available: Zyda, FineWeb, DCLM, and Dolma. Models trained on Zyda-2 significantly outperform identical models trained on the… See the full description on the dataset page: https://huggingface.co/datasets/Zyphra/Zyda-2.
Intended Uses
This model is registered on the QuantaMrkt quantum-safe registry. This model has not yet been PQC-verified.
Quick Start
# Install the CLI pip install quantumshield # Pull the model quantumshield pull Zyphra/Zyda-2 # Verify file integrity quantumshield verify Zyphra/Zyda-2
About
Zyda-2 Zyda-2 is a 5 trillion token language modeling dataset created by collecting open and high quality datasets and combining them and cross-deduplication and model-based quality filtering. Zyda-2 comprises diverse sources of web data, highly educational content, math, code, and scientific papers. To construct Zyda-2, we took the best open-source datasets available: Zyda, FineWeb, DCLM, and Dolma. Models trained on Zyda-2 significantly outperform identical models trained on the… See the full description on the dataset page: https://huggingface.co/datasets/Zyphra/Zyda-2.
Get this model
Pull with QuantumShield
quantumshield pull Zyphra/Zyda-2 Verify signatures
quantumshield verify Zyphra/Zyda-2