ASTERIZER

REZIRETSA's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

updated a model 1 day ago

ASTERIZER/Ezaris-Instruct

Text Generation • Updated 1 day ago

updated a dataset 1 day ago

ASTERIZER/Ezaris-Training-Sets

Updated 1 day ago • 310

updated a collection 1 day ago

Ezaris Tokenizer

Collection

Multilingual BPE tokenizer family (128K/64K/8K) for LUNA-1B, with benchmark vs the original GPT-NeoX tokenizer. • 1 item • Updated 1 day ago

updated a dataset 1 day ago

ASTERIZER/Ezaris-Tokenizer

Updated 1 day ago • 72

updated a collection 1 day ago

Ezaris Tokenizer

Collection

Multilingual BPE tokenizer family (128K/64K/8K) for LUNA-1B, with benchmark vs the original GPT-NeoX tokenizer. • 1 item • Updated 1 day ago

updated a collection 2 days ago

Ezaris Program

Collection

LUNA-1B by ASTERIZER: 1B-parameter multilingual model, 2.4TB training corpus, and 128K/64K tokenizer. • 4 items • Updated 1 day ago

updated a Space 2 days ago

Running

Ezaris Tokenizer Training

🔤

Train a custom byte-level BPE tokenizer for multilingual models

published a model 2 days ago

ASTERIZER/Ezaris-Instruct

Text Generation • Updated 1 day ago

updated a Space 2 days ago

Running

ASTERIZER

🚀

published a Space 2 days ago

Running

ASTERIZER

🚀

updated 2 collections 2 days ago

Ezaris Tokenizer

Collection

Multilingual BPE tokenizer family (128K/64K/8K) for LUNA-1B, with benchmark vs the original GPT-NeoX tokenizer. • 1 item • Updated 1 day ago

LUNA Tokenizer

Collection

Shared GPT-NeoX tokenizer (vocab 50304) for LUNA-100M and LUNA-300M — the single tokenizer behind both models. • 1 item • Updated 2 days ago