WariHima
/

hourai2-90m-base-v1

Model card Files Files and versions

蓬莱2 90m base v1

architecture:

qwen3next

tarined by:

rtx3060 * 2h
training code in ./training_codes

tokenizer from ku-nlp/gpt2-small-japanese-char

dataset:

OmniAICreator/Japanese-Wikipedia-202506
use 60k rows 1 epoch

Downloads last month: 15

Safetensors

Model size

92.4M params

Tensor type

BF16

·

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for WariHima/hourai2-90m-base-v1

Finetunes

Dataset used to train WariHima/hourai2-90m-base-v1

Collection including WariHima/hourai2-90m-base-v1

hourai2 蓬莱弐

16 items • Updated 10 days ago