Public/koboldcpp
1
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2026-09-17 08:19:49 +02:00
Code Issues Packages Projects Releases Wiki Activity

Merge commit '4aced7a63156555911157d3002f9d3ddef4a1e55' into concedo_experimental

Browse Source 
# Conflicts:
#	.devops/cann.Dockerfile
#	.devops/cpu.Dockerfile
#	.devops/cuda.Dockerfile
#	.devops/intel.Dockerfile
#	.devops/musa.Dockerfile
#	.devops/rocm.Dockerfile
#	.devops/tools.sh
#	.devops/vulkan.Dockerfile
#	.github/workflows/build.yml
#	.github/workflows/release.yml
#	.gitignore
#	docs/ops.md
#	docs/ops/SYCL.csv
#	examples/batched/batched.cpp
#	examples/eval-callback/eval-callback.cpp
#	examples/gen-docs/gen-docs.cpp
#	examples/lookahead/lookahead.cpp
#	examples/lookup/lookup-create.cpp
#	examples/lookup/lookup-stats.cpp
#	examples/lookup/lookup.cpp
#	examples/model-conversion/scripts/causal/compare-logits.py
#	examples/model-conversion/scripts/causal/run-org-model.py
#	examples/model-conversion/scripts/utils/check-nmse.py
#	examples/parallel/parallel.cpp
#	examples/retrieval/retrieval.cpp
#	examples/save-load-state/save-load-state.cpp
#	examples/speculative-simple/speculative-simple.cpp
#	examples/speculative/speculative.cpp
#	examples/training/finetune.cpp
#	ggml/CMakeLists.txt
#	ggml/src/ggml-cann/ggml-cann.cpp
#	ggml/src/ggml-cpu/repack.cpp
#	ggml/src/ggml-sycl/common.hpp
#	ggml/src/ggml-sycl/convert.cpp
#	ggml/src/ggml-sycl/dequantize.hpp
#	ggml/src/ggml-sycl/dpct/helper.hpp
#	ggml/src/ggml-sycl/element_wise.cpp
#	ggml/src/ggml-sycl/element_wise.hpp
#	ggml/src/ggml-sycl/ggml-sycl.cpp
#	ggml/src/ggml-sycl/mmvq.cpp
#	ggml/src/ggml-sycl/pad.cpp
#	ggml/src/ggml-sycl/ssm_conv.cpp
#	ggml/src/ggml-sycl/vecdotq.hpp
#	pyrightconfig.json
#	scripts/sync-ggml.last
#	tests/test-arg-parser.cpp
#	tests/test-backend-ops.cpp
#	tools/cvector-generator/cvector-generator.cpp
#	tools/imatrix/imatrix.cpp
#	tools/mtmd/CMakeLists.txt
#	tools/mtmd/clip.cpp
#	tools/perplexity/perplexity.cpp
#	tools/server/README.md
This commit is contained in:
Concedo
2025-12-16 23:14:12 +08:00
parent e88bf41fdc 4aced7a631
commit 050a5b1f52
72 changed files with 4190 additions and 3086 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/llama-model.cpp
+15 -17
View File
@@ -773,6 +773,7 @@ void llama_model::load_hparams(llama_model_loader & ml) {
hparams.n_swa = 8192;
hparams.n_attn_temp_floor_scale = 8192;
hparams.f_attn_temp_scale = 0.1f;
hparams.f_attn_temp_offset = 1.0f;
hparams.set_swa_pattern(4); // pattern: 3 chunked - 1 full
}
@@ -1740,12 +1741,19 @@ void llama_model::load_hparams(llama_model_loader & ml) {
// that have no expert_gating_func model parameter set
hparams.expert_gating_func = LLAMA_EXPERT_GATING_FUNC_TYPE_SOFTMAX;
}
ml.get_key(LLM_KV_ROPE_SCALING_YARN_LOG_MUL, hparams.rope_yarn_log_mul, false);
if (ml.get_key(LLM_KV_ROPE_SCALING_YARN_LOG_MUL, hparams.rope_yarn_log_mul, 0.0f)) {
// [TAG_DEEPSEEK2_YARN_LOG_MUL_FIX]
// cancel the factor from the convert script
hparams.rope_yarn_log_mul /= 0.1f;
}
// (optional) temperature tuning - used by mistral-large
ml.get_key(LLM_KV_ATTENTION_TEMPERATURE_SCALE, hparams.f_attn_temp_scale, false);
ml.get_key(LLM_KV_ATTENTION_TEMPERATURE_LENGTH, hparams.n_attn_temp_floor_scale, false);
hparams.f_attn_temp_offset = 0.0f;
switch (hparams.n_layer) {
case 27: type = LLM_TYPE_16B; break;
case 60: type = LLM_TYPE_236B; break;
@@ -2372,9 +2380,11 @@ void llama_model::load_hparams(llama_model_loader & ml) {
ml.get_key(LLM_KV_ATTENTION_LAYERNORM_RMS_EPS, hparams.f_norm_rms_eps);
ml.get_key(LLM_KV_ATTENTION_TEMPERATURE_SCALE, hparams.f_attn_temp_scale, false);
ml.get_key(LLM_KV_ROPE_SCALING_YARN_BETA_FAST, hparams.yarn_beta_fast, false);
ml.get_key(LLM_KV_ROPE_SCALING_YARN_BETA_SLOW, hparams.yarn_beta_slow, false);
ml.get_key(LLM_KV_ROPE_SCALING_YARN_LOG_MUL, hparams.rope_yarn_log_mul, false);
ml.get_key(LLM_KV_ROPE_SCALING_YARN_BETA_FAST, hparams.yarn_beta_fast, false);
ml.get_key(LLM_KV_ROPE_SCALING_YARN_BETA_SLOW, hparams.yarn_beta_slow, false);
ml.get_key(LLM_KV_ROPE_SCALING_YARN_LOG_MUL, hparams.rope_yarn_log_mul, 0.0f);
hparams.f_attn_temp_offset = 0.0f;
// TODO: maybe add n_attn_temp_floor_scale as a separate KV?
if (hparams.f_attn_temp_scale != 0.0f) {
@@ -2384,18 +2394,6 @@ void llama_model::load_hparams(llama_model_loader & ml) {
}
}
// TODO: this seems to be correct with the case of mscale == mscale_all_dims == 1.0f
// but may need further verification with other values
if (hparams.rope_yarn_log_mul != 0.0f) {
float factor = 1.0f / hparams.rope_freq_scale_train;
float mscale = 1.0f;
float mscale_all_dims = hparams.rope_yarn_log_mul;
static auto get_mscale = [](float scale, float mscale) {
return scale <= 1.0f ? 1.0f : (0.1f * mscale * logf(scale) + 1.0f);
};
hparams.yarn_attn_factor = get_mscale(factor, mscale) / get_mscale(factor, mscale_all_dims);
}
switch (hparams.n_layer) {
case 26: type = LLM_TYPE_3B; break;
case 34: type = LLM_TYPE_8B; break;
@@ -6965,6 +6963,7 @@ void llama_model::print_info() const {
LLAMA_LOG_INFO("%s: freq_base_train = %.1f\n", __func__, hparams.rope_freq_base_train);
LLAMA_LOG_INFO("%s: freq_scale_train = %g\n", __func__, hparams.rope_freq_scale_train);
LLAMA_LOG_INFO("%s: n_ctx_orig_yarn = %u\n", __func__, hparams.n_ctx_orig_yarn);
LLAMA_LOG_INFO("%s: rope_yarn_log_mul= %.4f\n", __func__, hparams.rope_yarn_log_mul);
LLAMA_LOG_INFO("%s: rope_finetuned = %s\n", __func__, hparams.rope_finetuned ? "yes" : "unknown");
// MRoPE (Multi-axis Rotary Position Embedding) sections
if (const auto & s = hparams.rope_sections; s[0] || s[1] || s[2] || s[3]) {
@@ -7028,7 +7027,6 @@ void llama_model::print_info() const {
LLAMA_LOG_INFO("%s: expert_weights_scale = %.1f\n", __func__, hparams.expert_weights_scale);
LLAMA_LOG_INFO("%s: expert_weights_norm = %d\n", __func__, hparams.expert_weights_norm);
LLAMA_LOG_INFO("%s: expert_gating_func = %s\n", __func__, llama_expert_gating_func_name((llama_expert_gating_func_type) hparams.expert_gating_func));
LLAMA_LOG_INFO("%s: rope_yarn_log_mul = %.4f\n", __func__, hparams.rope_yarn_log_mul);
}
if (arch == LLM_ARCH_QWEN2MOE) {