Daniel Han
d8eb7b3a0c
Update _utils.py
2024-08-07 10:48:39 -07:00
Daniel Han
145e7dc882
Update _utils.py
2024-08-07 10:47:11 -07:00
Daniel Han
8818fa61f7
Update _utils.py
2024-08-07 01:11:06 -07:00
Daniel Han
785310af0b
Fix tokenizers ( #887 )
...
* Update pyproject.toml
* Update tokenizer_utils.py
* Update tokenizer_utils.py
* Update tokenizer_utils.py
* Update _utils.py
* Update _utils.py
* Update tokenizer_utils.py
* Update tokenizer_utils.py
* fix_tokenizer
* Update tokenizer_utils.py
* Update tokenizer_utils.py
2024-08-06 20:24:44 -07:00
Daniel Han
15b2072be2
Update README.md
2024-08-05 00:00:53 -07:00
Daniel Han
f2c0f8dd20
Update README.md
2024-08-04 23:59:57 -07:00
Daniel Han
ab47c809d7
Update README.md
2024-08-04 23:50:40 -07:00
Daniel Han
74d51905c7
Update llama.py
2024-08-04 23:49:35 -07:00
emuchogu
8b1fb471ea
pascal support ( #870 )
...
Co-authored-by: Edward Muchogu <muchogu@gmail.com>
2024-08-04 23:45:51 -07:00
moontidef
e3fa7aa908
fix: fix config.torch_dtype bug ( #874 )
...
fix the bug #404
and the bug https://github.com/hiyouga/LLaMA-Factory/issues/4698#issue-2393500878
2024-08-04 23:45:34 -07:00
Daniel Han
65d1330dc4
Update pyproject.toml
2024-08-04 11:28:21 -07:00
Daniel Han
fcc785f413
Merge branch 'main' into nightly
2024-08-01 18:19:38 -07:00
Daniel Han
c8cec943c6
Fix RoPE extension ( #846 )
...
* 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>
2024-07-31 12:10:33 -07:00
Daniel Han
b27c236875
Update gemma.py
2024-07-31 12:09:33 -07:00
Daniel Han
edda5a5983
Merge branch 'nightly' of https://github.com/unslothai/unsloth into nightly
2024-07-31 12:05:28 -07:00
XiaoYang
5967c57ad9
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>
2024-07-31 12:05:08 -07:00
Daniel Han
9f2dce500e
Merge branch 'main' into nightly
2024-07-31 12:05:01 -07:00
Daniel Han
4537b882bc
Update README.md
2024-07-31 09:50:11 -07:00
Daniel Han
b12a165204
Gemma ( #843 )
...
* 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
2024-07-31 08:54:58 -07:00
Daniel Han
b06da09252
Update _utils.py
2024-07-31 08:54:22 -07:00
Daniel Han
3c437bdb8d
Update README.md
2024-07-31 08:53:20 -07:00
Daniel Han
cd64b6f59b
Update mapper.py
2024-07-30 23:51:47 -07:00
Daniel Han
b6b511f151
Update gemma2.py
2024-07-30 23:11:35 -07:00
Daniel Han
410d88efc3
Update gemma2.py
2024-07-30 23:11:23 -07:00
Daniel Han
b76e502a02
Update gemma2.py
2024-07-30 23:07:47 -07:00
Daniel Han
42aa263254
Update gemma2.py
2024-07-30 23:04:00 -07:00
Daniel Han
9b9e742418
flash-attn softcapping
2024-07-30 22:48:53 -07:00
Daniel Han
454ec3d73c
Update _utils.py
2024-07-30 19:57:53 -07:00
Daniel Han
7cf9ef592a
bugs
2024-07-30 19:56:36 -07:00
Daniel Han
a722100b30
Update llama.py
2024-07-30 10:29:54 -07:00
Daniel Han
c8a1cff41d
Update loader.py
2024-07-30 10:18:51 -07:00
XiaoYang
538dbf2bad
fix UnboundLocalError ( #834 )
...
* 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>
2024-07-30 10:15:09 -07:00
Daniel Han
04e779e33e
Merge branch 'main' into nightly
2024-07-30 10:01:11 -07:00
Daniel Han
18b5575993
Better debugging ( #826 )
...
* Update __init__.py
* Edits
* Checks
* Update _utils.py
* Update _utils.py
* Update loader.py
* Update _utils.py
* Update mapper.py
* Update loader.py
* Update loader.py
* Update _utils.py
* Update loader.py
* Update loader.py
* Update loader.py
* Update loader.py
* Update loader.py
* Update loader.py
* Update mapper.py
* Update loader.py
2024-07-28 00:10:02 -07:00
Daniel Han
4bb1d3d89d
Update loader.py
2024-07-27 23:04:23 -07:00
Daniel Han
3843f0a226
Update mapper.py
2024-07-27 23:03:29 -07:00
Daniel Han
27c457149e
Update loader.py
2024-07-27 23:03:13 -07:00
Daniel Han
1c58f3f2ce
Update loader.py
2024-07-27 23:02:10 -07:00
Daniel Han
0a5c4b5f2f
Update loader.py
2024-07-27 23:00:47 -07:00
Daniel Han
dc693e1a20
Update loader.py
2024-07-27 22:59:47 -07:00
Daniel Han
1f8c053cca
Update loader.py
2024-07-27 22:59:16 -07:00
Daniel Han
b0689f5b3a
Update loader.py
2024-07-27 22:58:15 -07:00
Daniel Han
2a392dcac9
Update _utils.py
2024-07-27 22:56:59 -07:00
Daniel Han
5460ad1b94
Update loader.py
2024-07-27 22:54:49 -07:00
Daniel Han
b5d27d4c40
Update loader.py
2024-07-27 22:52:06 -07:00
Daniel Han
47477efbfd
Update mapper.py
2024-07-27 22:50:38 -07:00
Daniel Han
462cb70436
Update _utils.py
2024-07-27 22:48:52 -07:00
Daniel Han
106582d07c
Update loader.py
2024-07-27 22:35:24 -07:00
Daniel Han
6dfc30702f
Update _utils.py
2024-07-27 22:32:39 -07:00
Daniel Han
a5fa935f4a
Update _utils.py
2024-07-27 22:30:48 -07:00