transformers
Fix GlmMoeDsaConfig default mlp_layer_types in modular conversion
#43876
Merged

Commits
  • Fix GlmMoeDsaConfig default mlp layer pattern
    weiguangli-io committed 144 days ago
  • fix(glm-moe-dsa): dedupe config init and colocate test
    weiguangli-io committed 144 days ago
  • Apply repo consistency fixes
    github-actions[bot] committed 144 days ago
Loading