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 'c78e682245f856ab5cfc2ffc0f8c20e8e12f163f' into concedo_experimental

Browse Source 
# Conflicts:
#	src/models/qwen35.cpp
#	src/models/qwen35moe.cpp
This commit is contained in:
Concedo
2026-02-20 14:41:32 +08:00
parent 9eb9e4eb83 c78e682245
commit 07c45ced56
26 changed files with 981 additions and 2027 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/llama-context.cpp
+3 -44
View File
@@ -722,8 +722,6 @@ int64_t llama_context::output_resolve_row(int32_t i) const {
}
float * llama_context::get_logits_ith(int32_t i) {
int64_t j = -1;
output_reorder();
try {
@@ -731,26 +729,7 @@ float * llama_context::get_logits_ith(int32_t i) {
throw std::runtime_error("no logits");
}
// TODO: use output_resolve_row()
if (i < 0) {
j = n_outputs + i;
if (j < 0) {
throw std::runtime_error(format("negative index out of range [0, %d)", n_outputs));
}
} else if ((size_t) i >= output_ids.size()) {
throw std::runtime_error(format("out of range [0, %zu)", output_ids.size()));
} else {
j = output_ids[i];
}
if (j < 0) {
throw std::runtime_error(format("batch.logits[%d] != true", i));
}
if (j >= n_outputs) {
// This should not happen
throw std::runtime_error(format("corrupt output buffer (j=%" PRId64 ", n_outputs=%d)", j, n_outputs));
}
const int64_t j = output_resolve_row(i);
return logits.data + j*model.vocab.n_tokens();
} catch (const std::exception & err) {
LLAMA_LOG_ERROR("%s: invalid logits id %d, reason: %s\n", __func__, i, err.what());
@@ -773,8 +752,6 @@ llama_token * llama_context::get_sampled_tokens() const{
}
float * llama_context::get_embeddings_ith(int32_t i) {
int64_t j = -1;
output_reorder();
try {
@@ -782,26 +759,7 @@ float * llama_context::get_embeddings_ith(int32_t i) {
throw std::runtime_error("no embeddings");
}
// TODO: use output_resolve_row()
if (i < 0) {
j = n_outputs + i;
if (j < 0) {
throw std::runtime_error(format("negative index out of range [0, %d)", n_outputs));
}
} else if ((size_t) i >= output_ids.size()) {
throw std::runtime_error(format("out of range [0, %zu)", output_ids.size()));
} else {
j = output_ids[i];
}
if (j < 0) {
throw std::runtime_error(format("batch.logits[%d] != true", i));
}
if (j >= n_outputs) {
// This should not happen
throw std::runtime_error(format("corrupt output buffer (j=%" PRId64 ", n_outputs=%d)", j, n_outputs));
}
const int64_t j = output_resolve_row(i);
const uint32_t n_embd_out = model.hparams.n_embd_out();
return embd.data + j*n_embd_out;
} catch (const std::exception & err) {
@@ -2771,6 +2729,7 @@ void llama_context::opt_init(struct llama_model * model, struct llama_opt_params
llama_set_param(model->cls_b, param_filter, param_filter_ud);
llama_set_param(model->cls_out, param_filter, param_filter_ud);
llama_set_param(model->cls_out_b, param_filter, param_filter_ud);
llama_set_param(model->cls_norm, param_filter, param_filter_ud);
for (struct llama_layer & layer : model->layers) {
for (size_t i = 0; i < sizeof(layer)/sizeof(struct ggml_tensor *); ++i) {