thek0tyara/llama-cpp-turboquant
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 4

chore : correct typos [no ci] (#20041)

Browse source 
* fix(docs): correct typos found during code review

Non-functional changes only:
- Fixed minor spelling mistakes in comments
- Corrected typos in user-facing strings
- No variables, logic, or functional code was modified.

Signed-off-by: Marcel Petrick <mail@marcelpetrick.it>

* Update docs/backend/CANN.md

Co-authored-by: Aaron Teo <taronaeo@gmail.com>

* Revert "Auxiliary commit to revert individual files from 846d1c301281178efbc6ce6060ad34c1ebe45af8"

This reverts commit 02fcf0c7db661d5ff3eff96b2b2db9fdb7213256.

* Update tests/test-backend-ops.cpp

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

* Update tests/test-backend-ops.cpp

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

---------

Signed-off-by: Marcel Petrick <mail@marcelpetrick.it>
Co-authored-by: Aaron Teo <taronaeo@gmail.com>
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
This commit is contained in:
Marcel Petrick 2026-03-05 08:50:21 +01:00 • committed by GitHub
parent 7a99dc85e2
commit 92f7da00b4
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
81 changed files with 160 additions and 160 deletions
Download patch file Download diff file Expand all files Collapse all files

12
convert_hf_to_gguf.py
View file

@ -4031,7 +4031,7 @@ class Qwen2VLVisionModel(MmprojModel):
# split Conv3D into Conv2Ds
c1, c2, kt, kh, kw = data_torch.shape
del c1, c2, kh, kw # unused
assert kt == 2, "Current implmentation only support temporal_patch_size of 2"
assert kt == 2, "Current implementation only support temporal_patch_size of 2"
yield (gguf.TENSOR_NAMES[gguf.MODEL_TENSOR.V_ENC_EMBD_PATCH] + ".weight" , data_torch[:, :, 0, ...])
yield (gguf.TENSOR_NAMES[gguf.MODEL_TENSOR.V_ENC_EMBD_PATCH] + ".weight.1", data_torch[:, :, 1, ...])
else:
@ -5404,7 +5404,7 @@ class KimiLinearModel(TextModel):
# Get ssm_d_conv from linear_attn_config.short_conv_kernel_size or ssm_d_conv
linear_attn_config = self.hparams["linear_attn_config"]
# n_head == 0 for KDA layers, n_head > 0 for MLA layers
# full_attention_layers list will be used to distingush layer type
# full_attention_layers list will be used to distinguish layer type
_num_kv_heads = list()
_full_attn_layers = linear_attn_config["full_attn_layers"]
for il in range(self.hparams["num_hidden_layers"]):
@ -6505,7 +6505,7 @@ class Gemma3VisionModel(MmprojModel):
super().set_gguf_parameters()
hparams = self.hparams
self.gguf_writer.add_clip_projector_type(gguf.VisionProjectorType.GEMMA3)
# default values below are taken from HF tranformers code
# default values below are taken from HF transformers code
self.gguf_writer.add_vision_attention_layernorm_eps(hparams.get("layer_norm_eps", 1e-6))
self.gguf_writer.add_vision_use_gelu(True)
# calculate proj_scale_factor (used by tinygemma3 test model)
@ -7097,7 +7097,7 @@ class Rwkv7Model(TextModel):
if bid == 0 and "time_mix_a" in new_name:
# dummy v0/v1/v2 on first layer
# easist way to make llama happy
# easiest way to make llama happy
yield (new_name.replace("time_mix_a", "time_mix_v"), data_torch)
yield (new_name, data_torch)
@ -9596,7 +9596,7 @@ class GraniteHybridModel(Mamba2Model, GraniteMoeModel):
# NOTE: Explicitly include hparam prefix prefix for d_model to
# disambiguate with top-level head_dim
# NOTE 2: If needed for future models, this can be isolated in a method
# to separate the prefix setting and teh keys used
# to separate the prefix setting and the keys used
self.d_model = self.find_hparam([f"{self.hparam_prefixes[0]}_head_dim", "hidden_size", "d_model"])
self.n_group = self.find_hparam(["n_groups", "num_groups"])
self.d_inner = self.find_hparam(["expand", "num_heads"]) * self.d_model
@ -9743,7 +9743,7 @@ class NemotronHModel(GraniteHybridModel):
self.gguf_writer.add_value_length(self.head_dim)
# Set feed_forward_length
# NOTE: This will trigger an override warning. This is preferrable to
# NOTE: This will trigger an override warning. This is preferable to
# duplicating all the parent logic
if not self.is_moe:
n_ff = self.find_hparam(["intermediate_size", "n_inner", "hidden_dim"])