Commit History

Upload folder using huggingface_hub
17163ae
Running
verified

yashvyasop commited on

Update Blog.md
629b1ef
verified

yashvyasop commited on

Upload folder using huggingface_hub
322a982
verified

yashvyasop commited on

Upload folder using huggingface_hub
3edbdb5
verified

yashvyasop commited on

Upload folder using huggingface_hub
b30ff25
verified

yashvyasop commited on

Upload folder using huggingface_hub
9b34b37
verified

yashvyasop commited on

grpo: add --eval_best_of (best-of-N eval for SFT and GRPO)
8a2235b
verified

yashvyasop commited on

grpo: add --eval_best_of (best-of-N eval for SFT and GRPO)
97a5555
verified

yashvyasop commited on

GRPO v3_anchored: fix promote mode-collapse
e10110e
verified

yashvyasop commited on

GRPO v2_delta_phase reward + diversified states + tuned smoke schedule
da8eeea
verified

yashvyasop commited on

GRPO v2_delta_phase reward + diversified states + tuned smoke schedule
715e7e0
verified

yashvyasop commited on

Colab: HF Jobs orchestrator (full 400/200/3 run)
9dcee98
verified

yashvyasop commited on

Update usage comment for full training
04a98b0
verified

yashvyasop commited on

Bump default max_steps to 200
fe2d531
verified

yashvyasop commited on

Fix notebook: unsloth-first imports, HF spaces clone, GPU check
86e410d
verified

yashvyasop commited on

Add Colab notebook for full GRPO training
009f50c
verified

yashvyasop commited on

Patch missing warnings_issued attr for transformers 5.x compat
9eb8e28
verified

yashvyasop commited on

Tighten pydantic pin to <2.11 to match mergekit ~=2.10.6
1994102
verified

yashvyasop commited on

Pin pydantic <2.12 after editable install to fix mergekit compat
80f9fa1
verified

yashvyasop commited on

Add weave dep (another TRL 0.24 hard import in callbacks.py)
fdd1260
verified

yashvyasop commited on

Patch TRANSFORMERS_CACHE in sanity check too
d29d073
verified

yashvyasop commited on

Patch TRANSFORMERS_CACHE for llm_blender compat
83ee735
verified

yashvyasop commited on

Add llm-blender dep required by trl 0.24 judges
fad96c9
verified

yashvyasop commited on

Add mergekit dep required by trl 0.24 callbacks
6ad440d
verified

yashvyasop commited on

Fix total_mem -> total_memory in GPU check
d24665d
verified

yashvyasop commited on

Add GRPO shell launcher with pinned deps
fcdfb56
verified

yashvyasop commited on

Add standalone GRPO training script
81863b2
verified

yashvyasop commited on

Fix GRPO smoke dependencies with llm-blender
f63b625
verified

yashvyasop commited on

Fix GRPO smoke dependencies with llm-blender
8ef8d11
verified

yashvyasop commited on

Fix GRPO training script import path
a8c1549
verified

yashvyasop commited on

Add PYTHONPATH check to GRPO smoke launcher
2bd1cf1
verified

yashvyasop commited on

Fix GRPO training script import path
04307dd
verified

yashvyasop commited on

Fix GRPO smoke dependencies with llm-blender
bb75a43
verified

yashvyasop commited on

Pin TRL version for GRPO smoke job
b675479
verified

yashvyasop commited on

Fix GRPO smoke job by installing mergekit
9b04641
verified

yashvyasop commited on

Fix GRPO smoke job torchao incompatibility
d4fe404
verified

yashvyasop commited on

Add GRPO smoke job launcher
9c5dd45
verified

yashvyasop commited on

Add GRPO smoke job launcher
a4c81e5
verified

yashvyasop commited on

Add GRPO training script for DesignGym 2.0
df2f172
verified

yashvyasop commited on

Upload folder using huggingface_hub
05acbb9
verified

yashvyasop commited on

Upload folder using huggingface_hub
ec45357
verified

yashvyasop commited on

Upload folder using huggingface_hub
7b90e91
verified

yashvyasop commited on

Upload folder using huggingface_hub
44c2d9e
verified

yashvyasop commited on

Upload folder using huggingface_hub
f48180b
verified

yashvyasop commited on

Upload folder using huggingface_hub
1b32a8c
verified

yashvyasop commited on

Upload folder using huggingface_hub
f95e1d0
verified

yashvyasop commited on

Upload folder using huggingface_hub
566fbb4
verified

yashvyasop commited on

Upload folder using huggingface_hub
73f77c2
verified

yashvyasop commited on

Upload folder using huggingface_hub
4ee1718
verified

yashvyasop commited on

initial commit
8137a06
verified

yashvyasop commited on