Public/koboldcpp
1
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2026-09-10 06:49:12 +02:00
Code Issues Packages Projects Releases Wiki Activity

note: clip_is_mrope was moved to mtmd_decode_use_mrope upstream and no longer syncs since https://github.com/ggml-org/llama.cpp/pull/18793

Browse Source 
Merge commit 'c1e79e610fd28f2c3923539fee9313734bbf8cfa' into concedo_experimental

# Conflicts:
#	.github/workflows/build.yml
#	.github/workflows/release.yml
#	CMakeLists.txt
#	CONTRIBUTING.md
#	MIT_LICENSE_GGML_SDCPP_LLAMACPP_ONLY.md
#	README.md
#	SECURITY.md
#	ci/run.sh
#	common/CMakeLists.txt
#	common/arg.cpp
#	docs/ops.md
#	docs/ops/BLAS.csv
#	docs/ops/zDNN.csv
#	docs/preset.md
#	examples/batched/batched.cpp
#	examples/debug/debug.cpp
#	ggml/src/ggml-blas/CMakeLists.txt
#	ggml/src/ggml-opencl/CMakeLists.txt
#	ggml/src/ggml-opencl/ggml-opencl.cpp
#	licenses/LICENSE-curl
#	licenses/LICENSE-httplib
#	scripts/pr2wt.sh
#	scripts/sync_vendor.py
#	tests/CMakeLists.txt
#	tests/test-backend-ops.cpp
#	tools/cli/README.md
#	tools/completion/README.md
#	tools/llama-bench/llama-bench.cpp
#	tools/server/README.md
#	vendor/cpp-httplib/LICENSE
This commit is contained in:
Concedo
2026-01-13 23:31:14 +08:00
parent 0dc18c668c c1e79e610f
commit 7d2c1c4f46
45 changed files with 859 additions and 661 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/llama-model.cpp
+4 -1
View File
@@ -6925,7 +6925,10 @@ bool llama_model::load_tensors(llama_model_loader & ml) {
} else {
// Linear attention (gated delta net) specific tensors
// Create tensors with calculated dimensions
layer.ssm_in = create_tensor(tn(LLM_TENSOR_SSM_IN, "weight", i), { n_embd, qkvz_dim }, 0);
// note: ssm_in is used by legacy GGUF
layer.ssm_in = create_tensor(tn(LLM_TENSOR_SSM_IN, "weight", i), { n_embd, qkvz_dim }, TENSOR_NOT_REQUIRED);
layer.wqkv = create_tensor(tn(LLM_TENSOR_ATTN_QKV, "weight", i), { n_embd, key_dim * 2 + value_dim }, TENSOR_NOT_REQUIRED);
layer.wqkv_gate = create_tensor(tn(LLM_TENSOR_ATTN_GATE, "weight", i), { n_embd, value_dim }, TENSOR_NOT_REQUIRED);
layer.ssm_conv1d = create_tensor(tn(LLM_TENSOR_SSM_CONV1D, "weight", i), { hparams.ssm_d_conv, conv_dim }, 0);
layer.ssm_dt = create_tensor(tn(LLM_TENSOR_SSM_DT, "bias", i), { hparams.ssm_dt_rank }, 0);
layer.ssm_a = create_tensor(tn(LLM_TENSOR_SSM_A_NOSCAN, i), { hparams.ssm_dt_rank }, 0);