Additional training on Japanese data by CyberAgent for deepseek-r1.
14b
32b
2,191 Pulls Updated 3 months ago
f4d24e9138dd · 148B
{
"stop": [
"<|begin▁of▁sentence|>",
"<|end▁of▁sentence|>",
"<|User|>",
"<|Assistant|>"
]
}