Search results

we need more code data

Data

openhands

Status: Open.

#81 In FlagAI-Open/OpenSeek;

· zacliu2023 opened on Mar 13, 2025

Question: Multi-stage RL worth trying?

questionFurther information is requestedFurther information is requested

Status: Open.

#50 In FlagAI-Open/OpenSeek;

· fangyuan-ksgk opened on Feb 26, 2025

Architecture Improvement: Self-Attention part

enhancementNew feature or requestNew feature or request

Infra

Status: Open.

Feature

#48 In FlagAI-Open/OpenSeek;

· LoserCheems opened on Feb 24, 2025

Architecture Improvement: Residual part

enhancementNew feature or requestNew feature or request

Infra

Status: Open.

Feature

#47 In FlagAI-Open/OpenSeek;

· LoserCheems opened on Feb 24, 2025

Architecture Improvement: Feedforward network part

enhancementNew feature or requestNew feature or request

Infra

Status: Open.

Feature

#45 In FlagAI-Open/OpenSeek;

· LoserCheems opened on Feb 20, 2025

Will we release a R1-like model for code generation, specifically for competitive programming？

questionFurther information is requestedFurther information is requested

Status: Open.

Feature

#44 In FlagAI-Open/OpenSeek;

· phonism opened on Feb 20, 2025

Distributed Training System: Visualization of the communication flows and scheduling relationships in complex large-scale distributed clusters

enhancementNew feature or requestNew feature or request

Infra

Status: Open.

#42 In FlagAI-Open/OpenSeek;

· aoyulong opened on Feb 20, 2025

·  

OpenSeek-stage2

Distributed Training System: Implement more innovative FP8 training solutions

enhancementNew feature or requestNew feature or request

Infra

Status: Open.

Feature

#41 In FlagAI-Open/OpenSeek;

· aoyulong opened on Feb 20, 2025

·  

OpenSeek-stage2

Distributed Training System: Implement customized parallel and optimization strategies for specific chips

enhancementNew feature or requestNew feature or request

Infra

Status: Open.

#40 In FlagAI-Open/OpenSeek;

· aoyulong opened on Feb 20, 2025

·  

OpenSeek-stage2

Distributed Training System: Collaborate with algorithm teams to achieve more efficient long sequence optimization

enhancementNew feature or requestNew feature or request

Infra

Status: Open.

Feature

#39 In FlagAI-Open/OpenSeek;

· aoyulong opened on Feb 20, 2025

·  

OpenSeek-stage2

Distributed Training System: Improve communication implementation to achieve more efficient MoE parallelism optimization

enhancementNew feature or requestNew feature or request

Infra

Status: Open.

#38 In FlagAI-Open/OpenSeek;

· aoyulong opened on Feb 20, 2025

·  

OpenSeek-stage2

Distributed Training System: Achieve more efficient pipeline parallelism

enhancementNew feature or requestNew feature or request

Infra

Status: Open.

#37 In FlagAI-Open/OpenSeek;

· aoyulong opened on Feb 20, 2025

·  

OpenSeek-stage2