Spaces:
Sleeping
Sleeping
Commit History
-
Upload folder using huggingface_hub
Verified signature1e65f1e -
Upload folder using huggingface_hub
Verified signaturefa2d96c
-
Upload folder using huggingface_hub
Verified signature17163ae
-
Update Blog.md
Verified signature629b1ef -
Upload folder using huggingface_hub
Verified signature322a982 -
Upload folder using huggingface_hub
Verified signature3edbdb5 -
Upload folder using huggingface_hub
Verified signatureb30ff25 -
Upload folder using huggingface_hub
Verified signature9b34b37 -
grpo: add --eval_best_of (best-of-N eval for SFT and GRPO)
Verified signature8a2235b -
grpo: add --eval_best_of (best-of-N eval for SFT and GRPO)
Verified signature97a5555 -
GRPO v3_anchored: fix promote mode-collapse
Verified signaturee10110e -
GRPO v2_delta_phase reward + diversified states + tuned smoke schedule
Verified signatureda8eeea -
GRPO v2_delta_phase reward + diversified states + tuned smoke schedule
Verified signature715e7e0 -
Colab: HF Jobs orchestrator (full 400/200/3 run)
Verified signature9dcee98 -
Update usage comment for full training
Verified signature04a98b0 -
Bump default max_steps to 200
Verified signaturefe2d531 -
Fix notebook: unsloth-first imports, HF spaces clone, GPU check
Verified signature86e410d -
Add Colab notebook for full GRPO training
Verified signature009f50c -
Patch missing warnings_issued attr for transformers 5.x compat
Verified signature9eb8e28 -
Tighten pydantic pin to <2.11 to match mergekit ~=2.10.6
Verified signature1994102 -
Pin pydantic <2.12 after editable install to fix mergekit compat
Verified signature80f9fa1 -
Add weave dep (another TRL 0.24 hard import in callbacks.py)
Verified signaturefdd1260 -
Patch TRANSFORMERS_CACHE in sanity check too
Verified signatured29d073 -
Patch TRANSFORMERS_CACHE for llm_blender compat
Verified signature83ee735 -
Add llm-blender dep required by trl 0.24 judges
Verified signaturefad96c9 -
Add mergekit dep required by trl 0.24 callbacks
Verified signature6ad440d -
Fix total_mem -> total_memory in GPU check
Verified signatured24665d -
Add GRPO shell launcher with pinned deps
Verified signaturefcdfb56 -
Add standalone GRPO training script
Verified signature81863b2 -
Fix GRPO smoke dependencies with llm-blender
Verified signaturef63b625 -
Fix GRPO smoke dependencies with llm-blender
Verified signature8ef8d11 -
Fix GRPO training script import path
Verified signaturea8c1549 -
Add PYTHONPATH check to GRPO smoke launcher
Verified signature2bd1cf1 -
Fix GRPO training script import path
Verified signature04307dd -
Fix GRPO smoke dependencies with llm-blender
Verified signaturebb75a43 -
Pin TRL version for GRPO smoke job
Verified signatureb675479 -
Fix GRPO smoke job by installing mergekit
Verified signature9b04641 -
Fix GRPO smoke job torchao incompatibility
Verified signatured4fe404 -
Add GRPO smoke job launcher
Verified signature9c5dd45 -
Add GRPO smoke job launcher
Verified signaturea4c81e5
-
Add GRPO training script for DesignGym 2.0
Verified signaturedf2f172 -
Upload folder using huggingface_hub
Verified signature05acbb9 -
Upload folder using huggingface_hub
Verified signatureec45357 -
Upload folder using huggingface_hub
Verified signature7b90e91 -
Upload folder using huggingface_hub
Verified signature44c2d9e
-
Upload folder using huggingface_hub
Verified signaturef48180b -
Upload folder using huggingface_hub
Verified signature1b32a8c -
Upload folder using huggingface_hub
Verified signaturef95e1d0 -
Upload folder using huggingface_hub
Verified signature566fbb4 -
Upload folder using huggingface_hub
Verified signature73f77c2