Skip to content

fix model name delivery and log more info #1745

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

Open
Harold-lkk wants to merge 7 commits into
base: agent_dev
Choose a base branch
from
Open

fix model name delivery and log more info #1745

Show file tree
Hide file tree
Changes from 4 commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
9f9242e
Revert "Support lmdeploy routed experts shared-store keys (#1701)"
Harold-lkk Apr 29, 2026
afb856d
【Fix】R3 MultiTurn
Harold-lkk Apr 29, 2026
b0f7b57
fix model name delivery and log more info
Harold-lkk Apr 30, 2026
b698bdd
support routed export store
Harold-lkk May 1, 2026
9f2d61e
fix r3 release and log daemon
Harold-lkk May 2, 2026
0dc51e7
update config
Harold-lkk May 6, 2026
5a748e8
return r3 with np array
Harold-lkk May 6, 2026
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
12 changes: 9 additions & 3 deletions examples/v1/config/agent_rl_qwen35_30b_grpo.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -48,7 +48,7 @@
max_concurrent_groups = 512

max_prompt_length = 4096
pack_max_length = 68 * 1024
pack_max_length = 256 * 1024
max_response_length = 64 * 1024

train_ep_size = 1
Expand All @@ -66,7 +66,7 @@
lr = 1e-6
hf_interval = 5
total_epochs = 10
sp_size = 1
sp_size = 4
# evaluation settings
enable_evaluate = False
enable_initial_evaluate = False
Expand Down Expand Up @@ -218,8 +218,14 @@ def convert_rollout_tractory_to_train(env, group_data_items):
model_cfg.text_config.z_loss_cfg = None
model_cfg.text_config.balancing_loss_cfg = None
model_cfg.text_config.freeze_routers = True
# model_cfg.text_config.mtp_config = MTPConfig(
# num_layers=1,
# loss_scaling_factor=1.0,
# detach_mtp_lm_head_weight=True,
# detach_mtp_inputs=True,
# share_weights=False,
# )
model_cfg.text_config.vocab_size = 251392
# model_cfg.text_config.embed_grad_max_token_id = 251173

optim_cfg = AdamWConfig(
lr=lr,
Expand Down
Loading