vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

fix: change name

Browse source
This commit is contained in:
younesbelkada 2024-08-18 14:42:40 +00:00
parent 4553502d4a
commit d637bb97bc
2 changed files with 4 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

2
convert_hf_to_gguf.py
View file

@ -2759,7 +2759,7 @@ class MambaModel(Model):
self.gguf_writer.add_ssm_state_size(d_state)
self.gguf_writer.add_ssm_time_step_rank(dt_rank)
self.gguf_writer.add_layer_norm_rms_eps(rms_norm_eps)
self.gguf_writer.add_mamba_dt_b_c_rms(use_dt_b_c_norm) # For classic Mamba we don't apply rms norm on B / DT layers
self.gguf_writer.add_ssm_dt_b_c_rms(use_dt_b_c_norm) # For classic Mamba we don't apply rms norm on B / DT layers
self.gguf_writer.add_file_type(self.ftype)
_tok_embd = None

6
gguf-py/gguf/gguf_writer.py
View file

@ -714,9 +714,6 @@ class GGUFWriter:
def add_rope_scaling_finetuned(self, value: bool) -> None:
self.add_bool(Keys.Rope.SCALING_FINETUNED.format(arch=self.arch), value)
def add_ssm_dt_b_c_rms(self, value: bool) -> None:
self.add_bool(Keys.SSM.DT_B_C_RMS.format(arch=self.arch), value)
def add_rope_scaling_yarn_log_mul(self, value: float) -> None:
self.add_float32(Keys.Rope.SCALING_YARN_LOG_MUL.format(arch=self.arch), value)
@ -733,6 +730,9 @@ class GGUFWriter:
def add_ssm_time_step_rank(self, value: int) -> None:
self.add_uint32(Keys.SSM.TIME_STEP_RANK.format(arch=self.arch), value)
def add_ssm_dt_b_c_rms(self, value: bool) -> None:
self.add_bool(Keys.SSM.DT_B_C_RMS.format(arch=self.arch), value)
def add_tokenizer_model(self, model: str) -> None:
self.add_string(Keys.Tokenizer.MODEL, model)