Downloads · 30 days
4
10% of all-time downloads
lilyzhng/gpt-oss-20b-tb2-grpo-lora
gpt-oss-20b-tb2-grpo-lora is a machine learning model from lilyzhng. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for peft.
LoRA after GRPO on Nemotron probe band (rewardmode=ctrfonly).
Downloads · 30 days
4
10% of all-time downloads
All-time downloads
42
Public
Repo size
63.7 MB
Likes
0
Public
Click a slice to open those files.
.safetensors63.7 MB · 100%
From the Hugging Face model README
LoRA after GRPO on Nemotron probe band (reward_mode=ctrf_only).
| Field | Value |
|---|---|
| Base model | openai/gpt-oss-20b |
| Init adapter | lilyzhng/gpt-oss-20b-tb2-sft-lora |
| Modal checkpoint | t33b-ctrf_only-train-13steps |
| LoRA rank | 32 |
vllm serve openai/gpt-oss-20b --enable-lora --enable-mixed-moe-lora-format \
--lora-modules '{"name":"tb2-grpo","path":"lilyzhng/gpt-oss-20b-tb2-grpo-lora","is_3d_lora_weight":false}' \
--trust-remote-code