Merge branch 'ml-explore:main' into adding-reporting-to-wandb

This commit is contained in:
Gökdeniz Gülmez
2025-03-12 14:35:39 +01:00
committed by GitHub
17 changed files with 561 additions and 139 deletions

View File

@@ -98,6 +98,7 @@ def linear_to_lora_layers(
"minicpm",
"deepseek",
"olmo2",
"olmoe",
"internlm3",
]:
keys = set(["self_attn.q_proj", "self_attn.v_proj"])
@@ -106,6 +107,8 @@ def linear_to_lora_layers(
if model.model_type == "qwen2_moe":
keys.add("mlp.gate")
keys.add("mlp.shared_expert_gate")
if model.model_type == "olmoe":
keys.add("mlp.gate")
elif model.model_type == "gpt_bigcode":
keys = set(["attn.c_attn"])