Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
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
5 changes: 0 additions & 5 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -241,8 +241,3 @@ peft_model = PeftModel.from_pretrained(model, 'Qwen3.5-4B-LoRA')
## 🏛 License

This framework is licensed under the [Apache License (Version 2.0)](https://github.com/modelscope/mcore-bridge/blob/master/LICENSE). For models and datasets, please refer to the original resource page and follow the corresponding License.


## Star History

[![Star History Chart](https://api.star-history.com/svg?repos=modelscope/mcore-bridge&type=Date)](https://star-history.com/#modelscope/mcore-bridge&Date)
5 changes: 0 additions & 5 deletions README_zh.md
Original file line number Diff line number Diff line change
Expand Up @@ -238,8 +238,3 @@ peft_model = PeftModel.from_pretrained(model, 'Qwen3.5-4B-LoRA')
## 🏛 License

本框架使用[Apache License (Version 2.0)](https://github.com/modelscope/mcore-bridge/blob/master/LICENSE)进行许可。模型和数据集请查看原资源页面并遵守对应License。


## Star History

[![Star History Chart](https://api.star-history.com/svg?repos=modelscope/mcore-bridge&type=Date)](https://star-history.com/#modelscope/mcore-bridge&Date)
3 changes: 1 addition & 2 deletions src/mcore_bridge/patcher.py
Original file line number Diff line number Diff line change
Expand Up @@ -689,8 +689,7 @@ def forward_with_scores(
try:
from megatron.core.transformer.experimental_attention_variant.dsa import fused_qk_topk_naive
except ImportError:
raise ImportError('fused_qk_topk_naive is not available. Please install megatron-core from source. '
'`pip install git+https://github.com/NVIDIA/Megatron-LM.git`')
raise ImportError('fused_qk_topk_naive is not available. Please install "megatron-core>=0.17.0"')
# [seqlen, batch, index_n_heads * index_head_dim]
# [seqlen, batch, index_head_dim]
# [seqlen, batch, index_n_heads]
Expand Down
Loading