Commit graph

1457 commits

Author SHA1 Message Date
pre-commit-ci[bot]
60fb6cae11 [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
2026-02-20 04:58:47 +00:00
Jai Suphavadeeprasit
ccdd5a1ca6 linting 2026-02-19 23:57:47 -05:00
Jai Suphavadeeprasit
c1a80205cc update readme 2026-02-19 23:50:05 -05:00
Jai Suphavadeeprasit
11fabaa7f7 gsm8k trial 2026-02-19 21:55:33 -05:00
Jai Suphavadeeprasit
809b88bf30 gsm8k trial 2026-02-19 21:32:40 -05:00
Jai Suphavadeeprasit
bbbfaf1680 gsm8k trial 2026-02-19 21:17:49 -05:00
Jai Suphavadeeprasit
0dcc9156d2 change OPD style 2026-02-19 19:19:23 -05:00
Jai Suphavadeeprasit
527433b5bc change OPD style 2026-02-19 17:08:27 -05:00
Jai Suphavadeeprasit
33f5696171 Merge branch 'pipelineRL' into OnPolicyDistillation 2026-02-19 16:39:21 -05:00
Jai Suphavadeeprasit
01f090af0d script test 2026-02-19 16:07:26 -05:00
Jai Suphavadeeprasit
0b266758db script test 2026-02-19 16:00:50 -05:00
Jai Suphavadeeprasit
ef9f29dbde readme fix 2026-02-19 15:49:29 -05:00
Jai Suphavadeeprasit
657945fa1d sanity_check 2026-02-19 15:18:32 -05:00
Jai Suphavadeeprasit
00908ec366 packageification 2026-02-19 15:16:24 -05:00
Jai Suphavadeeprasit
d438702ee4 packageification 2026-02-19 13:37:33 -05:00
Jai Suphavadeeprasit
301d0b2699 model layer stuff 2026-02-18 10:52:20 -05:00
Jai Suphavadeeprasit
fae3f5b09e readme fixes 2026-02-17 13:44:48 -05:00
Jai Suphavadeeprasit
366ea72384 readme fixes 2026-02-17 12:24:07 -05:00
Jai Suphavadeeprasit
bc0f9ee625 debug changes 2026-02-17 08:15:07 -05:00
Jai Suphavadeeprasit
f52de7441c found bug 2026-02-16 21:26:44 -05:00
Jai Suphavadeeprasit
573221497d base env debugging 2026-02-16 21:23:54 -05:00
Jai Suphavadeeprasit
7a90f34d85 base env debugging 2026-02-16 21:20:33 -05:00
Jai Suphavadeeprasit
b0658f6327 base env debugging 2026-02-16 21:05:57 -05:00
Jai Suphavadeeprasit
0e81c62e90 on policy changes 2026-02-16 17:39:37 -05:00
Jai Suphavadeeprasit
becadb54b0 Fix math_server_zero.py to support CLI OpenAI arguments
Change ServerBaseline to APIServerConfig in config_init() so that
--openai.base_url and other CLI arguments work for on-policy distillation.

Co-authored-by: Cursor <cursoragent@cursor.com>
2026-02-16 17:18:01 -05:00
Jai Suphavadeeprasit
cc9b891eba initial commit 2026-02-16 11:46:20 -05:00
Jai Suphavadeeprasit
43cc71e070 cleanup 3 2026-02-13 12:39:37 -05:00
Jai Suphavadeeprasit
39d307b440 cleanup 2 2026-02-13 11:56:55 -05:00
Jai Suphavadeeprasit
da0583f47d cleanup 2026-02-13 11:26:26 -05:00
Jai Suphavadeeprasit
95de25aa37 restart issues 3 2026-02-13 11:26:26 -05:00
Jai Suphavadeeprasit
813ac83195 restart issues 3 2026-02-13 11:26:26 -05:00
Jai Suphavadeeprasit
8ee92651b4 restart issues 2 2026-02-13 11:26:26 -05:00
Jai Suphavadeeprasit
e0f77d43d0 restart issues 2 2026-02-13 11:26:26 -05:00
Jai Suphavadeeprasit
0fab717427 restart issues 2026-02-13 11:26:26 -05:00
Jai Suphavadeeprasit
2169943066 math zero 32k 2026-02-13 11:26:26 -05:00
Jai Suphavadeeprasit
6f8ebc99fe math zero 32k 2026-02-13 11:26:26 -05:00
Jai Suphavadeeprasit
e0d5c989be math zero 2026-02-13 11:26:26 -05:00
Jai Suphavadeeprasit
f88a00331e kill old 2026-02-13 11:26:26 -05:00
Jai Suphavadeeprasit
33844c374b gradient flow fix 2026-02-13 11:26:26 -05:00
Jai Suphavadeeprasit
1c8bb34bc1 wandb integration 2026-02-13 11:26:26 -05:00
Jai Suphavadeeprasit
36ecce6ccb vllm restart 2 2026-02-13 11:26:26 -05:00
Jai Suphavadeeprasit
26cc509164 vllm restart 1 2026-02-13 11:26:26 -05:00
Jai Suphavadeeprasit
0f0a3f8afc vllm restart 2026-02-13 11:26:25 -05:00
Jai Suphavadeeprasit
6d4d705271 enforce eager check 32k context length 2026-02-13 11:26:25 -05:00
Jai Suphavadeeprasit
f37de35195 enforce eager check 2026-02-13 11:26:25 -05:00
Jai Suphavadeeprasit
6ce875afbe visibility fix 2026-02-13 11:26:25 -05:00
Jai Suphavadeeprasit
a6138b3c65 lora restart saving gradient changes 2026-02-13 11:26:25 -05:00
Jai Suphavadeeprasit
aa0c5792df ditching lora nccl 2 2026-02-13 11:26:25 -05:00
Jai Suphavadeeprasit
9ba6c0e7bb ditching lora nccl 2026-02-13 11:26:25 -05:00
Jai Suphavadeeprasit
28bf3d9d60 testing lora 2026-02-13 11:26:25 -05:00