admin/OpenClaw-RL
1
0
Fork 0
mirror of https://github.com/Gen-Verse/OpenClaw-RL.git synced 2026-08-14 00:47:53 +00:00
Code Issues Packages Projects Releases Wiki Activity

Update README.md

Browse Source
This commit is contained in:
Yinjie Wang
2026-04-21 00:15:33 -05:00
committed by GitHub
parent bec16108b1
commit 2035e75040
1 changed files with 7 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
toolcall-rl/README.md
+7
View File
@@ -82,6 +82,13 @@ PYTHONPATH=/path/to/Megatron-LM python tools/convert_hf_to_torch_dist.py \
--rotary-base 5000000 \
--save /path/to/qwen3-4b-sft_torch_dist
# or for qwen3.5:
source scripts/models/qwen3-4B.sh
PYTHONPATH=/path/to/Megatron-LM python tools/convert_hf_to_torch_dist.py \
${MODEL_ARGS[@]} \
--hf-checkpoint /path/to/qwen3.5-4b \
--save /path/to/qwen3.5-4b-torch_dist
# Download RL data and run
huggingface-cli download --repo-type dataset BytedTsinghua-SIA/DAPO-Math-17k --local-dir /path/to/dapo-math-17k
huggingface-cli download --repo-type dataset zhuzilin/aime-2024 --local-dir /path/to/aime-2024