Skip to content

(docs): Update README.md and fix typo in GRPO #149

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
PanAndy merged 1 commit into from
Aug 20, 2025
Merged

(docs): Update README.md and fix typo in GRPO #149

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 3 additions & 4 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -84,10 +84,9 @@ Leveraging a multi-role distributed architecture with Ray for flexible resource
[RAFT++](https://alibaba.github.io/ROLL/docs/English/UserGuide/algorithms/RAFT_Plus_Plus)
[StarPO](https://alibaba.github.io/ROLL/docs/English/UserGuide/algorithms/agentic_StarPO)

#### Beckend
#### Backend
[DeepSeed](https://alibaba.github.io/ROLL/docs/English/UserGuide/backend/deepspeed)
[Megatron](https://alibaba.github.io/ROLL/docs/English/UserGuide/backend/megatron)
[LoRA](https://alibaba.github.io/ROLL/docs/English/UserGuide/backend/lora)
[Megatron](https://alibaba.github.io/ROLL/docs/English/UserGuide/backend/megatron)
[vLLM](https://alibaba.github.io/ROLL/docs/English/UserGuide/backend/vllm)
[SGLang](https://alibaba.github.io/ROLL/docs/English/UserGuide/backend/sglang)

Expand Down Expand Up @@ -119,7 +118,7 @@ Leveraging a multi-role distributed architecture with Ray for flexible resource
* Inference/Generation supports vLLM, SGLang.
* Training supports DeepSpeed (ZeRO), Megatron-LM 5D parallelism (mcore-adapter, dp/tp/pp/cp/ep), FSDP under implementation.
* Extreme offload/reload capabilities.
* Supports LoRA training.
* Supports [LoRA](https://alibaba.github.io/ROLL/docs/English/UserGuide/backend/lora) training.
* Supports FP8 rollout (FP8 inference for LLM as judge, FP8 rollout with BF16 training under development).
* **AutoDeviceMapping:** Supports custom device mapping for different roles, flexibly managing colocated and disaggregated deployments.
* **Observability:** Integrated with SwanLab / WandB / TensorBoard, tracking of performance for each domain and reward type.
Expand Down
2 changes: 1 addition & 1 deletion docs_roll/docs/English/UserGuide/algorithms/GRPO.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -26,7 +26,7 @@ adv_estimator: "grpo"
ppo_epochs: 1
use_kl_loss: true
kl_loss_coef: 0.001
loss_agg_mode: "seq-mean-token-sum"
loss_agg_mode: "seq-mean-token-mean"

# ppo related
# advantage
Expand Down
2 changes: 1 addition & 1 deletion docs_roll/docs/简体中文/使用指南/algorithms/GRPO.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -26,7 +26,7 @@ adv_estimator: "grpo"
ppo_epochs: 1
use_kl_loss: true
kl_loss_coef: 0.001
loss_agg_mode: "seq-mean-token-sum"
loss_agg_mode: "seq-mean-token-mean"

# ppo related
# advantage
Expand Down