CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-11-18 11:46:58 +00:00
Code Issues Packages Projects Releases Wiki Activity

model : add grok-2 support (#15539)

Browse Source 
* add grok-2 support

* type fix

* type fix

* type fix

* "fix" vocab for invalid sequences

* fix expert tensor mapping and spaces in vocab

* add chat template

* fix norm tensor mapping

* rename layer_out_norm to ffn_post_norm

* ensure ffn_post_norm is mapped

* fix experts merging

* remove erroneous FFN_GATE entry

* concatenate split tensors and add more metadata

* process all expert layers and try cat instead of hstack

* add support for community BPE vocab

* fix expert feed forward length and ffn_down concat

* commit this too

* add ffn_up/gate/down, unsure if sequence is right

* add ffn_gate/down/up to tensor names

* correct residual moe (still not working)

* mess--

* fix embedding scale being applied twice

* add built in chat template

* change beta fast for grok if default value

* remove spm vocab in favor of community bpe vocab

* change attention temp length metadata type to integer

* update attention temp length metadata

* remove comment

* replace M_SQRT2 with std::sqrt(2)

* add yarn metadata, move defaults to hparams
This commit is contained in:
Sigbjørn Skjæret
2025-09-14 23:00:59 +02:00
committed by GitHub
parent 6c019cb04e
commit b8e09f08b9
16 changed files with 281 additions and 96 deletions
Download Patch File Download Diff File Expand all files Collapse all files

21
gguf-py/gguf/gguf_writer.py
View File

@@ -733,6 +733,9 @@ class GGUFWriter:
def add_attn_logit_softcapping(self, value: float) -> None:
self.add_float32(Keys.LLM.ATTN_LOGIT_SOFTCAPPING.format(arch=self.arch), value)
def add_router_logit_softcapping(self, value: float) -> None:
self.add_float32(Keys.LLM.ROUTER_LOGIT_SOFTCAPPING.format(arch=self.arch), value)
def add_final_logit_softcapping(self, value: float) -> None:
self.add_float32(Keys.LLM.FINAL_LOGIT_SOFTCAPPING.format(arch=self.arch), value)
@@ -829,6 +832,12 @@ class GGUFWriter:
def add_attention_scale(self, value: float) -> None:
self.add_float32(Keys.Attention.SCALE.format(arch=self.arch), value)
def add_attn_output_scale(self, value: float) -> None:
self.add_float32(Keys.Attention.OUTPUT_SCALE.format(arch=self.arch), value)
def add_attn_temperature_length(self, value: int) -> None:
self.add_uint32(Keys.Attention.TEMPERATURE_LENGTH.format(arch=self.arch), value)
def add_pooling_type(self, value: PoolingType) -> None:
self.add_uint32(Keys.LLM.POOLING_TYPE.format(arch=self.arch), value.value)
@@ -859,6 +868,18 @@ class GGUFWriter:
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)
def add_rope_scaling_yarn_ext_factor(self, value: float) -> None:
self.add_float32(Keys.Rope.SCALING_YARN_EXT_FACTOR.format(arch=self.arch), value)
def add_rope_scaling_yarn_attn_factor(self, value: float) -> None:
self.add_float32(Keys.Rope.SCALING_YARN_ATTN_FACTOR.format(arch=self.arch), value)
def add_rope_scaling_yarn_beta_fast(self, value: float) -> None:
self.add_float32(Keys.Rope.SCALING_YARN_BETA_FAST.format(arch=self.arch), value)
def add_rope_scaling_yarn_beta_slow(self, value: float) -> None:
self.add_float32(Keys.Rope.SCALING_YARN_BETA_SLOW.format(arch=self.arch), value)
def add_ssm_conv_kernel(self, value: int) -> None:
self.add_uint32(Keys.SSM.CONV_KERNEL.format(arch=self.arch), value)