CyberAgentLM2 is a decoder-only language model pre-trained on the 1.3T tokens of publicly available Japanese and English datasets.

Updated 2 years ago

2 years ago

71cf6b48923c · 4.3GB ·

archllama

parameters7.01B

quantizationQ4_K_M

4.3GB

{ "stop": [ "<|endoftext|>" ] }

36B

Readme

CyberAgentLM2 is a decoder-only language model pre-trained on the 1.3T tokens of publicly available Japanese and English datasets.

Instruct:

ollama run cyberagent-calm2:7b-chat

Text completion:

ollama run cyberagent-calm2:7b