Skip to content
Merged
Changes from 1 commit
Commits
Show all changes
285 commits
Select commit Hold shift + click to select a range
564b6f8
Upcast layernorms
danielhanchen Aug 17, 2025
b8a34b4
Update llama.py
danielhanchen Aug 17, 2025
509fcb5
Update llama.py
danielhanchen Aug 17, 2025
27f1a2e
Update llama.py
danielhanchen Aug 18, 2025
931851a
Update llama.py
danielhanchen Aug 18, 2025
3b9057b
Update llama.py
danielhanchen Aug 18, 2025
3dd87bb
Update llama.py
danielhanchen Aug 18, 2025
f3f2b51
Merge branch 'main' into nightly
danielhanchen Aug 18, 2025
b757faf
Update save.py
danielhanchen Aug 18, 2025
2e86333
Update rl.py
danielhanchen Aug 18, 2025
b01e948
Update pyproject.toml
danielhanchen Aug 18, 2025
b064255
Merge branch 'main' into nightly
danielhanchen Aug 18, 2025
a751fd7
Update rl.py
danielhanchen Aug 18, 2025
c5d22e1
Merge branch 'main' into nightly
danielhanchen Aug 18, 2025
3cb6eaf
Update rl_replacements.py
danielhanchen Aug 18, 2025
de77a26
Update rl.py
danielhanchen Aug 19, 2025
27ca531
Update rl.py
danielhanchen Aug 19, 2025
6514c8e
Update rl.py
danielhanchen Aug 19, 2025
3e29ae7
Update _utils.py
danielhanchen Aug 19, 2025
a42f624
Update __init__.py
danielhanchen Aug 19, 2025
9437f9e
Torch 2.8
danielhanchen Aug 19, 2025
1dd99a2
Update rl_replacements.py
danielhanchen Aug 19, 2025
5d5ece0
Merge branch 'main' into nightly
danielhanchen Aug 19, 2025
ecd8d38
Merge branch 'main' into nightly
danielhanchen Aug 19, 2025
89b5603
Merge branch 'main' into nightly
danielhanchen Aug 19, 2025
fa68976
Merge branch 'main' into nightly
danielhanchen Aug 20, 2025
5349cd0
Update loader.py
danielhanchen Aug 20, 2025
5a344c2
UNSLOTH_ENABLE_CCE
danielhanchen Aug 20, 2025
e56363c
Fix
danielhanchen Aug 20, 2025
c79aece
Update loader.py
danielhanchen Aug 20, 2025
c4b530c
Update loader.py
danielhanchen Aug 20, 2025
0913b58
Update __init__.py
danielhanchen Aug 20, 2025
374f703
Update __init__.py
danielhanchen Aug 20, 2025
c0efbec
Update __init__.py
danielhanchen Aug 20, 2025
761a445
Update __init__.py
danielhanchen Aug 20, 2025
30ea44c
Import fixes
danielhanchen Aug 20, 2025
c45467c
Update loader.py
danielhanchen Aug 20, 2025
55e4c78
Fix aimv2 issue
danielhanchen Aug 20, 2025
a160e42
Update loader.py
danielhanchen Aug 20, 2025
675c4ef
Update import_fixes.py
danielhanchen Aug 20, 2025
a99d6b2
Update import_fixes.py
danielhanchen Aug 20, 2025
7e82623
Update loader.py
danielhanchen Aug 20, 2025
0e678d6
Update loader.py
danielhanchen Aug 20, 2025
9b82317
Update loader.py
danielhanchen Aug 20, 2025
8a76fd3
Upgrade
danielhanchen Aug 20, 2025
94bcb28
Update loader.py
danielhanchen Aug 20, 2025
7d7a115
Update loader.py
danielhanchen Aug 20, 2025
031f5e1
Update loader.py
danielhanchen Aug 20, 2025
98bee64
Update loader.py
danielhanchen Aug 20, 2025
21fa9fd
Merge branch 'main' into nightly
danielhanchen Aug 20, 2025
2ba9008
Update vision.py
danielhanchen Aug 21, 2025
ea435e6
Update vision.py
danielhanchen Aug 21, 2025
5bebfa9
custom_datatype
danielhanchen Aug 21, 2025
356789a
recheck
danielhanchen Aug 21, 2025
d0f97a9
Float16
danielhanchen Aug 21, 2025
d83767f
Update vision.py
danielhanchen Aug 21, 2025
5b575d8
Update vision.py
danielhanchen Aug 21, 2025
66eee4d
Update vision.py
danielhanchen Aug 21, 2025
27d044e
Update vision.py
danielhanchen Aug 21, 2025
34d07d8
Update vision.py
danielhanchen Aug 21, 2025
3ad7561
Update loader.py
danielhanchen Aug 21, 2025
b757297
Update loader.py
danielhanchen Aug 21, 2025
ceeca86
Update loader.py
danielhanchen Aug 21, 2025
87758b9
Update loader.py
danielhanchen Aug 21, 2025
97d34d4
Update loader.py
danielhanchen Aug 21, 2025
43bf41f
Update loader.py
danielhanchen Aug 21, 2025
6e7ad52
Update loader.py
danielhanchen Aug 21, 2025
d605aa7
Update loader.py
danielhanchen Aug 21, 2025
f417dc8
Update loader.py
danielhanchen Aug 21, 2025
05fe3d1
Update loader.py
danielhanchen Aug 21, 2025
a79d6f6
Update loader.py
danielhanchen Aug 21, 2025
59702c4
Update loader.py
danielhanchen Aug 21, 2025
1b66aee
Update loader.py
danielhanchen Aug 21, 2025
a71fa05
Update loader.py
danielhanchen Aug 21, 2025
d3e8625
Update loader.py
danielhanchen Aug 21, 2025
fb112cf
Update loader.py
danielhanchen Aug 21, 2025
5dbdcc5
Update loader.py
danielhanchen Aug 21, 2025
fdaa007
Update loader.py
danielhanchen Aug 21, 2025
ba0eb04
Bug fix
danielhanchen Aug 21, 2025
3f98262
Update loader.py
danielhanchen Aug 21, 2025
3e6511b
Update loader.py
danielhanchen Aug 21, 2025
c9e7537
Update loader.py
danielhanchen Aug 21, 2025
2e38e8a
Update loader.py
danielhanchen Aug 22, 2025
8b3a8ba
Update loader.py
danielhanchen Aug 22, 2025
f706d20
torch_dtype
danielhanchen Aug 22, 2025
bf863a8
Merge branch 'main' into nightly
danielhanchen Aug 28, 2025
84ca61f
Merge branch 'main' into nightly
danielhanchen Aug 30, 2025
e82fd70
Merge branch 'main' into nightly
danielhanchen Sep 4, 2025
c61a21d
Merge branch 'main' into nightly
danielhanchen Sep 4, 2025
b56cc1b
Update rl.py
danielhanchen Sep 4, 2025
c47f936
Fix CE Loss
danielhanchen Sep 4, 2025
6093c4c
Merge branch 'main' into nightly
danielhanchen Sep 4, 2025
0b896c5
Versioning
danielhanchen Sep 4, 2025
327f517
Merge branch 'main' into nightly
danielhanchen Sep 4, 2025
5b0c47a
Merge branch 'main' into nightly
danielhanchen Sep 8, 2025
de5c3b5
Merge branch 'main' into nightly
danielhanchen Sep 9, 2025
7234a62
Update loader.py
danielhanchen Sep 9, 2025
68c1aba
Update loader.py
danielhanchen Sep 9, 2025
d07b819
Merge branch 'main' into nightly
danielhanchen Sep 9, 2025
05fc2f2
extract_model_type_from_config
danielhanchen Sep 9, 2025
99c7afb
Model types
danielhanchen Sep 10, 2025
fc5d91d
Update loader.py
danielhanchen Sep 10, 2025
702a9ea
get_transformers_model_type
danielhanchen Sep 10, 2025
8ece4a6
Update loader.py
danielhanchen Sep 10, 2025
f3ac0e3
Update loader.py
danielhanchen Sep 10, 2025
d2b0d41
Update loader.py
danielhanchen Sep 10, 2025
e5920fe
Update rl.py
danielhanchen Sep 10, 2025
bf0367e
Update pyproject.toml
danielhanchen Sep 10, 2025
d2c2cc1
Update loader.py
danielhanchen Sep 10, 2025
337557c
Merge branch 'main' into nightly
danielhanchen Sep 10, 2025
b038d5d
Merge branch 'main' into nightly
danielhanchen Sep 10, 2025
39da8b4
Merge branch 'main' into nightly
danielhanchen Sep 13, 2025
35ca177
Update loader.py
danielhanchen Sep 13, 2025
2eaf868
Update loader.py
danielhanchen Sep 13, 2025
7c892e7
Update loader.py
danielhanchen Sep 13, 2025
72ff24c
Versioning
danielhanchen Sep 14, 2025
9654895
Merge branch 'main' into nightly
danielhanchen Sep 15, 2025
227842c
Update _utils.py
danielhanchen Sep 15, 2025
505ae67
Update _utils.py
danielhanchen Sep 15, 2025
80465dc
Update _utils.py
danielhanchen Sep 15, 2025
4150e08
Update _utils.py
danielhanchen Sep 15, 2025
27bae35
Merge branch 'main' into nightly
danielhanchen Sep 15, 2025
7d4bf8d
Merge branch 'main' into nightly
danielhanchen Sep 15, 2025
e1f981b
Merge branch 'main' into nightly
danielhanchen Sep 16, 2025
032c2c8
Update vision.py
danielhanchen Sep 16, 2025
b105aae
Update vision.py
danielhanchen Sep 16, 2025
400df38
Fix DataParallel
danielhanchen Sep 16, 2025
809a8b3
Update _utils.py
danielhanchen Sep 16, 2025
a5c7fa6
Merge branch 'main' into nightly
danielhanchen Sep 16, 2025
78627e5
Merge branch 'main' into nightly
danielhanchen Sep 17, 2025
3dcc091
Update rl.py
danielhanchen Sep 17, 2025
28b1d50
Update synthetic.py
danielhanchen Sep 17, 2025
de162d3
Update synthetic.py
danielhanchen Sep 17, 2025
a507a7d
Update synthetic.py
danielhanchen Sep 17, 2025
cda7263
Update synthetic.py
danielhanchen Sep 17, 2025
dd8ad92
Update synthetic.py
danielhanchen Sep 17, 2025
a725b98
Update synthetic.py
danielhanchen Sep 17, 2025
321f1a3
Update synthetic.py
danielhanchen Sep 17, 2025
357e501
Update synthetic.py
danielhanchen Sep 17, 2025
8a03656
Update synthetic.py
danielhanchen Sep 17, 2025
d7832d0
Update synthetic.py
danielhanchen Sep 17, 2025
84f5434
Update synthetic.py
danielhanchen Sep 17, 2025
17b2e98
Update synthetic.py
danielhanchen Sep 17, 2025
58f658e
Merge branch 'main' into nightly
danielhanchen Sep 17, 2025
5364138
Update mapper.py
danielhanchen Sep 17, 2025
8dbd008
Versioning
danielhanchen Sep 17, 2025
256f8fe
Merge branch 'main' into nightly
danielhanchen Sep 18, 2025
d7ca79f
Update loader.py
danielhanchen Sep 18, 2025
bb90785
Update loader.py
danielhanchen Sep 18, 2025
3289826
Update rl.py
danielhanchen Sep 18, 2025
a042114
Versioning
danielhanchen Sep 18, 2025
dfa91f7
Merge branch 'main' into nightly
danielhanchen Sep 18, 2025
ffa04dd
Update _utils.py
danielhanchen Sep 18, 2025
b365444
Fix auto_mapping
danielhanchen Sep 19, 2025
c60dfb0
Merge branch 'main' into nightly
danielhanchen Sep 19, 2025
bbb8252
Merge branch 'main' into nightly
danielhanchen Sep 19, 2025
f88e880
Merge branch 'main' into nightly
danielhanchen Sep 20, 2025
5ce7bf8
Update loader.py
danielhanchen Sep 20, 2025
755e6e2
Update loader.py
danielhanchen Sep 20, 2025
d01b8af
Update vision.py
danielhanchen Sep 20, 2025
d048d3a
Update vision.py
danielhanchen Sep 21, 2025
81ba78e
Update loader.py
danielhanchen Sep 21, 2025
0bb74fe
Message
danielhanchen Sep 21, 2025
14fdb22
Update vision.py
danielhanchen Sep 21, 2025
ce4f2b6
Update loader.py
danielhanchen Sep 21, 2025
e333b03
Update vision.py
danielhanchen Sep 21, 2025
456d225
cache_implementation
danielhanchen Sep 21, 2025
1cd7b85
Update vision.py
danielhanchen Sep 21, 2025
2b0d219
Update loader.py
danielhanchen Sep 21, 2025
d1c9283
Update vision.py
danielhanchen Sep 21, 2025
a0df6ab
Update vision.py
danielhanchen Sep 21, 2025
450b2da
Update vision.py
danielhanchen Sep 21, 2025
b1116d5
Update loader.py
danielhanchen Sep 21, 2025
7210cb1
Update vision.py
danielhanchen Sep 21, 2025
f148170
Save max_seq_length
danielhanchen Sep 21, 2025
7fa66da
Update _utils.py
danielhanchen Sep 21, 2025
0b49db1
Update rl.py
danielhanchen Sep 22, 2025
f1c47f8
Update vision.py
danielhanchen Sep 22, 2025
27f6203
Update llama.py
danielhanchen Sep 22, 2025
f06179f
Mistral3 vllm (#3349)
Datta0 Sep 22, 2025
67a544d
Set padding to 0
danielhanchen Sep 22, 2025
7238327
Fix patch
danielhanchen Sep 23, 2025
8a1e6fb
fixup patch (#3359)
danielhanchen Sep 23, 2025
f0ec1ae
Update vision.py
danielhanchen Sep 23, 2025
a64a3b2
Versioning
danielhanchen Sep 23, 2025
1b7640b
Update vision.py
danielhanchen Sep 23, 2025
f5c4385
Update vision.py
danielhanchen Sep 24, 2025
8438a76
Update vision.py
danielhanchen Sep 24, 2025
5867273
Update vision.py
danielhanchen Sep 24, 2025
7b2bef1
Update vision.py
danielhanchen Sep 24, 2025
82a7697
Update vision.py
danielhanchen Sep 24, 2025
aa9b200
Update vision.py
danielhanchen Sep 24, 2025
eb1df23
Update vision.py
danielhanchen Sep 24, 2025
563aa35
Update vision.py
danielhanchen Sep 24, 2025
4bfde2e
Update vision.py
danielhanchen Sep 24, 2025
d6beafe
MXFP4 dequant
danielhanchen Sep 24, 2025
19cfe1b
Update loader.py
danielhanchen Sep 24, 2025
63a7f65
Update vision.py
danielhanchen Sep 24, 2025
df5282b
load_in_16bit
danielhanchen Sep 24, 2025
e7174b1
Update vision.py
danielhanchen Sep 24, 2025
ffe5aca
Update vision.py
danielhanchen Sep 24, 2025
81356cc
Update vision.py
danielhanchen Sep 24, 2025
2313ea9
Update rl.py
danielhanchen Sep 25, 2025
0c18d86
Update vision.py
danielhanchen Sep 26, 2025
19017fd
offload_embedding
danielhanchen Sep 26, 2025
77fca79
Update vision.py
danielhanchen Sep 26, 2025
92084ba
Update vision.py
danielhanchen Sep 26, 2025
499f939
Update vision.py
danielhanchen Sep 26, 2025
402af41
Merge branch 'main' into nightly
danielhanchen Sep 26, 2025
07723f5
Merge branch 'main' into nightly
danielhanchen Sep 26, 2025
7a499e4
Merge branch 'main' into nightly
danielhanchen Sep 26, 2025
fffcea8
Merge branch 'main' into nightly
danielhanchen Sep 28, 2025
f72c0a9
Update vision.py
danielhanchen Sep 28, 2025
2a7cfa0
Update vision.py
danielhanchen Sep 28, 2025
2577d81
Update vision.py
danielhanchen Sep 28, 2025
1eee987
Update rl_replacements.py
danielhanchen Sep 30, 2025
e7f3170
Merge branch 'main' into nightly
danielhanchen Sep 30, 2025
1edc796
Update loader.py
danielhanchen Sep 30, 2025
205d09c
Fix padding issue
danielhanchen Sep 30, 2025
07cc6ed
Update pyproject.toml
danielhanchen Sep 30, 2025
d225f7f
Update _utils.py
danielhanchen Sep 30, 2025
5d6c3d9
Update pyproject.toml
danielhanchen Sep 30, 2025
af56af3
Update _utils.py
danielhanchen Sep 30, 2025
ad080bb
Merge branch 'main' into nightly
danielhanchen Oct 1, 2025
eb2d403
Update vision.py
danielhanchen Oct 1, 2025
9bc76e8
Update vision.py
danielhanchen Oct 1, 2025
a0425bb
Update vision.py
danielhanchen Oct 1, 2025
b0ba73c
Update vision.py
danielhanchen Oct 1, 2025
f85a91a
Update vision.py
danielhanchen Oct 1, 2025
80dce6b
Merge branch 'main' into nightly
danielhanchen Oct 5, 2025
47f2ef7
Update vision.py
danielhanchen Oct 5, 2025
06fc86f
New models
danielhanchen Oct 5, 2025
dcf22c8
Merge branch 'main' into nightly
danielhanchen Oct 5, 2025
ca3426a
Merge branch 'main' into nightly
danielhanchen Oct 14, 2025
8a09a7d
Merge branch 'main' into nightly
danielhanchen Oct 14, 2025
20b9202
Merge branch 'main' into nightly
danielhanchen Oct 16, 2025
778da7d
Update llama.py
danielhanchen Oct 16, 2025
ed443ee
Versioning
danielhanchen Oct 16, 2025
da00e2f
Update _utils.py
danielhanchen Oct 16, 2025
250ea60
Update llama.py
danielhanchen Oct 16, 2025
a921ea6
Update _utils.py
danielhanchen Oct 16, 2025
f1e76eb
Merge branch 'main' into nightly
danielhanchen Oct 16, 2025
c90df87
Update llama.py
danielhanchen Oct 16, 2025
c64f011
Fix AMD
danielhanchen Oct 16, 2025
8eecf7d
Update _utils.py
danielhanchen Oct 16, 2025
c22b9a3
Update llama.py
danielhanchen Oct 16, 2025
38b9e00
Update vision.py
danielhanchen Oct 16, 2025
b99dcd5
DEVICE_TYPE_TORCH
danielhanchen Oct 16, 2025
19bc977
Update __init__.py
danielhanchen Oct 16, 2025
5aa6a39
Update __init__.py
danielhanchen Oct 16, 2025
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Next Next commit
Update rl.py
  • Loading branch information
danielhanchen committed Aug 19, 2025
commit 6514c8ee55baf15360f5bf840dcaf6e8cf9eeb0f
2 changes: 1 addition & 1 deletion unsloth/models/rl.py
Original file line number Diff line number Diff line change
Expand Up @@ -541,7 +541,7 @@ def _patch_trl_rl_trainers(trainer_file = "grpo_trainer"):
max_seq_length_pre = \
"""max_seq_length : Optional[int] = field(
default = None,
metadata = {{'help': 'Maximum sequence length to truncate to.'}},
metadata = {'help': 'Maximum sequence length to truncate to.'},
)"""
max_seq_length_call = "max_seq_length = max_seq_length,"
max_seq_length_post = "self.max_seq_length = max_seq_length"
Expand Down