CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-11-06 09:46:50 +00:00
Code Issues Packages Projects Releases Wiki Activity

d_ssm -> d_inner;

Browse Source
This commit is contained in:
younesbelkada
2025-07-08 10:18:43 +04:00
parent d2f46f18ac
commit 7d7da0b37e
6 changed files with 19 additions and 27 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
convert_hf_to_gguf.py
View File

@@ -6674,7 +6674,8 @@ class FalconH1Model(Mamba2Model):
# Add Falcon Mamba2 specific configuration
self.gguf_writer.add_uint32("falcon_h1.attention.head_dim", self.hparams["head_dim"])
self.gguf_writer.add_uint32("falcon_h1.ssm.mamba_d_ssm", self.hparams["mamba_d_ssm"])
self.gguf_writer.add_uint32("falcon_h1.ssm.mamba_d_inner", self.hparams["mamba_d_ssm"])
self.gguf_writer.add_ssm_inner_size(self.hparams["mamba_d_ssm"])
self.gguf_writer.add_uint32("falcon_h1.num_attention_heads", self.find_hparam(["num_attention_heads"]))
self.gguf_writer.add_uint32("falcon_h1.num_key_value_heads",
self.find_hparam(["num_key_value_heads"], optional=True) or