CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-11-17 11:37:10 +00:00
Code Issues Packages Projects Releases Wiki Activity

model : add AfmoeForCausalLM support (#16477)

Browse Source 
* Add AFMOE model support

* Update to vocab

* Add model sizing

* Undo Rope change for ARCEE model

* Address review comments

* Update modeling code is_sliding -> use_rope, replace hard-coded logic

* Fix AFMOE tokenizer

* Update convert_hf_to_gguf.py

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

* Update convert_hf_to_gguf.py

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

* Update AFMoE tokenizer class identification to be more unique

---------

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
This commit is contained in:
Bartowski
2025-11-14 07:54:10 -05:00
committed by GitHub
parent 6cd0cf72ce
commit e1fcf8b09b
14 changed files with 541 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
gguf-py/gguf/tensor_mapping.py
View File

@@ -314,6 +314,10 @@ class TensorNameMap:
"model.layers.{bid}.self_attn.sinks", # openai-moe
),
MODEL_TENSOR.ATTN_GATE: (
"model.layers.{bid}.self_attn.gate_proj", # afmoe
),
# Feed-forward norm
MODEL_TENSOR.FFN_NORM: (
"gpt_neox.layers.{bid}.post_attention_layernorm", # gptneox
@@ -340,11 +344,12 @@ class TensorNameMap:
"model.layers.{bid}.feedforward_layernorm", # apertus
),
# Post feed-forward norm
# Pre feed-forward norm
MODEL_TENSOR.FFN_PRE_NORM: (
"model.layers.{bid}.pre_feedforward_layernorm", # gemma2
"layers.{bid}.pre_feedforward_layernorm", # embeddinggemma
"model.layers.{bid}.pre_ff_layernorm.weight",
"model.layers.{bid}.pre_mlp_layernorm", # afmoe
),
# Post feed-forward norm
@@ -370,6 +375,7 @@ class TensorNameMap:
"model.layers.{bid}.mlp.gate.wg", # hunyuan
"model.layers.{bid}.block_sparse_moe.primary_router", # smallthinker
"model.layers.{bid}.feed_forward.gate", # lfm2moe
"model.layers.{bid}.mlp.router.gate", # afmoe
),
MODEL_TENSOR.FFN_GATE_INP_SHEXP: (
@@ -380,6 +386,7 @@ class TensorNameMap:
"model.layers.{bid}.mlp.gate.e_score_correction", # deepseek-v3 dots1
"model.layers.{bid}.mlp.moe_statics.e_score_correction", # ernie4.5-moe
"model.layers.{bid}.mlp.gate.expert_bias", # bailingmoe2
"model.layers.{bid}.mlp.expert_bias", # afmoe
"model.layers.{bid}.feed_forward.expert_bias", # lfm2moe
"model.layers.{bid}.block_sparse_moe.e_score_correction", # minimax-m2
),