Skip to content

[Conversion Mapping] Small fixups #45483

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

Merged
vasqu merged 4 commits into from
Apr 20, 2026
Merged

[Conversion Mapping] Small fixups #45483

Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
d995078
fix
vasqu Apr 16, 2026
bb08642
Merge branch 'main' into fix-conversion-mapping
vasqu Apr 17, 2026
5904edc
Merge branch 'main' into fix-conversion-mapping
vasqu Apr 20, 2026
97fde4f
fix
vasqu Apr 20, 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
120 changes: 46 additions & 74 deletions src/transformers/conversion_mapping.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -475,16 +475,6 @@ def _build_checkpoint_conversion_mapping():
operations=[MergeModulelist(dim=0)],
),
],
"legacy": [
WeightRenaming(
source_patterns="LayerNorm.gamma",
target_patterns="LayerNorm.weight",
),
WeightRenaming(
source_patterns="LayerNorm.beta",
target_patterns="LayerNorm.bias",
),
],
"nomic_bert": [
WeightRenaming(r"encoder.layers", r"layers"),
WeightRenaming(r"emb_ln", r"embeddings.LayerNorm"),
Expand Down Expand Up @@ -523,6 +513,38 @@ def _build_checkpoint_conversion_mapping():
WeightRenaming(source_patterns="norm1", target_patterns="post_attention_layernorm"),
WeightRenaming(source_patterns="norm2", target_patterns="post_mlp_layernorm"),
],
"cohere_asr": [
WeightRenaming(r"encoder\.pre_encode\.conv\.", r"encoder.subsampling.layers."),
WeightRenaming(r"encoder\.pre_encode\.out\.", r"encoder.subsampling.linear."),
WeightRenaming(r"transf_decoder\._embedding\.position_embedding\.pos_enc", r"decoder.pos_emb.weight"),
WeightRenaming(r"transf_decoder\._embedding\.token_embedding", r"decoder.embed_tokens"),
WeightRenaming(r"transf_decoder\._embedding\.layer_norm", r"decoder.embedding_layernorm"),
WeightRenaming(r"transf_decoder\._decoder\.final_layer_norm", r"decoder.norm"),
WeightRenaming(r"transf_decoder\._decoder\.layers", r"decoder.layers"),
WeightRenaming(r"encoder_decoder_proj\.", r"decoder.proj."),
WeightRenaming(r"encoder\.(.+)\.self_attn\.linear_q", r"encoder.\1.self_attn.q_proj"),
WeightRenaming(r"encoder\.(.+)\.self_attn\.linear_k", r"encoder.\1.self_attn.k_proj"),
WeightRenaming(r"encoder\.(.+)\.self_attn\.linear_v", r"encoder.\1.self_attn.v_proj"),
WeightRenaming(r"encoder\.(.+)\.self_attn\.linear_out", r"encoder.\1.self_attn.o_proj"),
WeightRenaming(r"encoder\.(.+)\.self_attn\.linear_pos", r"encoder.\1.self_attn.relative_k_proj"),
WeightRenaming(r"encoder\.(.+)\.self_attn\.pos_bias_u", r"encoder.\1.self_attn.bias_u"),
WeightRenaming(r"encoder\.(.+)\.self_attn\.pos_bias_v", r"encoder.\1.self_attn.bias_v"),
WeightRenaming(r"decoder\.(.+)\.first_sub_layer\.query_net", r"decoder.\1.self_attn.q_proj"),
WeightRenaming(r"decoder\.(.+)\.first_sub_layer\.key_net", r"decoder.\1.self_attn.k_proj"),
WeightRenaming(r"decoder\.(.+)\.first_sub_layer\.value_net", r"decoder.\1.self_attn.v_proj"),
WeightRenaming(r"decoder\.(.+)\.first_sub_layer\.out_projection", r"decoder.\1.self_attn.o_proj"),
WeightRenaming(r"\.second_sub_layer\.query_net", r".encoder_attn.q_proj"),
WeightRenaming(r"\.second_sub_layer\.key_net", r".encoder_attn.k_proj"),
WeightRenaming(r"\.second_sub_layer\.value_net", r".encoder_attn.v_proj"),
WeightRenaming(r"\.second_sub_layer\.out_projection", r".encoder_attn.o_proj"),
WeightRenaming(r"\.third_sub_layer\.dense_in", r".mlp.fc1"),
WeightRenaming(r"\.third_sub_layer\.dense_out", r".mlp.fc2"),
WeightRenaming(r"\.layer_norm_1\.", r".input_layernorm."),
WeightRenaming(r"\.layer_norm_2\.", r".post_attention_layernorm."),
WeightRenaming(r"\.layer_norm_3\.", r".final_layernorm."),
WeightRenaming(r"\.conv\.batch_norm", r".conv.norm"),
WeightRenaming(r"log_softmax\.mlp\.layer0", r"proj_out"),
],
"qianfan_ocr": [
WeightRenaming(r"^vision_model\.", r"model\.vision_tower\."),
WeightRenaming(r"encoder\.layers\.", r"layers\."),
Expand Down Expand Up @@ -550,6 +572,16 @@ def _build_checkpoint_conversion_mapping():
WeightRenaming(r"^mlp1\.1\.", r"model\.multi_modal_projector\.linear_1\."),
WeightRenaming(r"^mlp1\.3\.", r"model\.multi_modal_projector\.linear_2\."),
],
"legacy": [
WeightRenaming(
source_patterns="LayerNorm.gamma",
target_patterns="LayerNorm.weight",
),
WeightRenaming(
source_patterns="LayerNorm.beta",
target_patterns="LayerNorm.bias",
),
],
}
# The legacy mapping is added to the esm model here since the extra weight renaming do not apply to the esm model.
mapping["esm"] += mapping["legacy"].copy()
Expand All @@ -565,78 +597,18 @@ def _build_checkpoint_conversion_mapping():
),
]

mapping["ernie4_5_moe"] = [
WeightRenaming("mlp.moe_statics.e_score_correction_bias", "mlp.gate.moe_statics.e_score_correction_bias"),
WeightConverter(
source_patterns=[
"mlp.experts.*.gate_proj.weight",
"mlp.experts.*.up_proj.weight",
],
target_patterns="mlp.experts.gate_up_proj",
operations=[MergeModulelist(dim=0), Concatenate(dim=1)],
),
WeightConverter(
source_patterns="mlp.experts.*.down_proj.weight",
target_patterns="mlp.experts.down_proj",
operations=[MergeModulelist(dim=0)],
),
mapping["ernie4_5_moe"] = mapping["qwen2_moe"].copy()
mapping["ernie4_5_moe"] += [
WeightRenaming("mlp.moe_statics.e_score_correction_bias", "mlp.gate.moe_statics.e_score_correction_bias")
]

mapping["minimax_m2"] = mapping["mixtral"].copy()
mapping["minimax_m2"] += [
WeightRenaming(".block_sparse_moe.e_score_correction_bias", ".mlp.e_score_correction_bias"),
]
mapping["exaone_moe"] = mapping["qwen2_moe"].copy()
mapping["exaone_moe"] += [WeightRenaming("mlp.e_score_correction_bias", "mlp.gate.e_score_correction_bias")]

mapping["solar_open"] = [
WeightConverter(
source_patterns=[
Comment on lines -591 to -593

@zucchini-nlp zucchini-nlp Apr 17, 2026

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

is it safe to delete? from PR description seems like mapping["solar_open"] = mapping["qwen2_moe"].copy()

@vasqu vasqu Apr 17, 2026

Copy link
Copy Markdown
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We have it already here

transformers/src/transformers/conversion_mapping.py

Line 51 in ff4f96a

"solar_open": "qwen2_moe",

Not sure why this was even recreated, not rushing to merge. Asked @3outeille whether this was really needed

@zucchini-nlp zucchini-nlp Apr 17, 2026

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

oh wow, didn't see that

"mlp.experts.*.gate_proj.weight",
"mlp.experts.*.up_proj.weight",
],
target_patterns="mlp.experts.gate_up_proj",
operations=[MergeModulelist(dim=0), Concatenate(dim=1)],
),
WeightConverter(
source_patterns="mlp.experts.*.down_proj.weight",
target_patterns="mlp.experts.down_proj",
operations=[MergeModulelist(dim=0)],
),
]

mapping["cohere_asr"] = [
WeightRenaming(r"encoder\.pre_encode\.conv\.", r"encoder.subsampling.layers."),
WeightRenaming(r"encoder\.pre_encode\.out\.", r"encoder.subsampling.linear."),
WeightRenaming(r"transf_decoder\._embedding\.position_embedding\.pos_enc", r"decoder.pos_emb.weight"),
WeightRenaming(r"transf_decoder\._embedding\.token_embedding", r"decoder.embed_tokens"),
WeightRenaming(r"transf_decoder\._embedding\.layer_norm", r"decoder.embedding_layernorm"),
WeightRenaming(r"transf_decoder\._decoder\.final_layer_norm", r"decoder.norm"),
WeightRenaming(r"transf_decoder\._decoder\.layers", r"decoder.layers"),
WeightRenaming(r"encoder_decoder_proj\.", r"decoder.proj."),
WeightRenaming(r"encoder\.(.+)\.self_attn\.linear_q", r"encoder.\1.self_attn.q_proj"),
WeightRenaming(r"encoder\.(.+)\.self_attn\.linear_k", r"encoder.\1.self_attn.k_proj"),
WeightRenaming(r"encoder\.(.+)\.self_attn\.linear_v", r"encoder.\1.self_attn.v_proj"),
WeightRenaming(r"encoder\.(.+)\.self_attn\.linear_out", r"encoder.\1.self_attn.o_proj"),
WeightRenaming(r"encoder\.(.+)\.self_attn\.linear_pos", r"encoder.\1.self_attn.relative_k_proj"),
WeightRenaming(r"encoder\.(.+)\.self_attn\.pos_bias_u", r"encoder.\1.self_attn.bias_u"),
WeightRenaming(r"encoder\.(.+)\.self_attn\.pos_bias_v", r"encoder.\1.self_attn.bias_v"),
WeightRenaming(r"decoder\.(.+)\.first_sub_layer\.query_net", r"decoder.\1.self_attn.q_proj"),
WeightRenaming(r"decoder\.(.+)\.first_sub_layer\.key_net", r"decoder.\1.self_attn.k_proj"),
WeightRenaming(r"decoder\.(.+)\.first_sub_layer\.value_net", r"decoder.\1.self_attn.v_proj"),
WeightRenaming(r"decoder\.(.+)\.first_sub_layer\.out_projection", r"decoder.\1.self_attn.o_proj"),
WeightRenaming(r"\.second_sub_layer\.query_net", r".encoder_attn.q_proj"),
WeightRenaming(r"\.second_sub_layer\.key_net", r".encoder_attn.k_proj"),
WeightRenaming(r"\.second_sub_layer\.value_net", r".encoder_attn.v_proj"),
WeightRenaming(r"\.second_sub_layer\.out_projection", r".encoder_attn.o_proj"),
WeightRenaming(r"\.third_sub_layer\.dense_in", r".mlp.fc1"),
WeightRenaming(r"\.third_sub_layer\.dense_out", r".mlp.fc2"),
WeightRenaming(r"\.layer_norm_1\.", r".input_layernorm."),
WeightRenaming(r"\.layer_norm_2\.", r".post_attention_layernorm."),
WeightRenaming(r"\.layer_norm_3\.", r".final_layernorm."),
WeightRenaming(r"\.conv\.batch_norm", r".conv.norm"),
WeightRenaming(r"log_softmax\.mlp\.layer0", r"proj_out"),
]

for model_type, base_pattern in _MODEL_TO_CONVERSION_PATTERN.items():
if model_type in mapping:
continue
Expand Down
Loading