* bugs
* Update _utils.py
* flash-attn softcapping
* Update gemma2.py
* Update gemma2.py
* Update gemma2.py
* Update gemma2.py
* Update mapper.py
* Update README.md
* Update _utils.py
* Fix ROPE extension issue and device mismatch (#840)
* When an exception has been assigned using as target, it is cleared at the end of the except clause.(https://docs.python.org/3/reference/compound_stmts.html#the-try-statement)
* Update loader.py
* round up to extend rope size
* inv_freq.device changed, make sure they are on the same device
---------
Co-authored-by: xiaoyang <xiaoyang@youzan.com>
Co-authored-by: Daniel Han <danielhanchen@gmail.com>
* Update gemma.py
---------
Co-authored-by: XiaoYang <xyangk@gmail.com>
Co-authored-by: xiaoyang <xiaoyang@youzan.com>
* When an exception has been assigned using as target, it is cleared at the end of the except clause.(https://docs.python.org/3/reference/compound_stmts.html#the-try-statement)
* Update loader.py
* round up to extend rope size
* inv_freq.device changed, make sure they are on the same device
---------
Co-authored-by: xiaoyang <xiaoyang@youzan.com>
Co-authored-by: Daniel Han <danielhanchen@gmail.com>
* When an exception has been assigned using as target, it is cleared at the end of the except clause.(https://docs.python.org/3/reference/compound_stmts.html#the-try-statement)
* Update loader.py
---------
Co-authored-by: xiaoyang <xiaoyang@youzan.com>
Co-authored-by: Daniel Han <danielhanchen@gmail.com>
