root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-12-26 03:14:35 +00:00
Code Issues Actions 7 Packages Projects Releases Wiki Activity

train : add general name (#6752)

Browse Source 
* llama : make general.name optional

* train: Add 'general.name' to model metadata

Signed-off-by: teleprint-me <77757836+teleprint-me@users.noreply.github.com>

---------

Signed-off-by: teleprint-me <77757836+teleprint-me@users.noreply.github.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
This commit is contained in:
Austin 2024-04-19 03:16:45 -04:00 committed by GitHub
parent bca40e9814
commit 8b1b1f4982
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 3 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
examples/train-text-from-scratch/train-text-from-scratch.cpp
View File

@ -73,6 +73,7 @@ struct my_llama_model {
static const char * LLM_KV_TRAINING_TYPE_TRAIN_MODEL = "train_model"; static const char * LLM_KV_TRAINING_TYPE_TRAIN_MODEL = "train_model";
static const char * LLM_KV_TRAINING_TYPE = "training.type"; static const char * LLM_KV_TRAINING_TYPE = "training.type";
static const char * LLM_KV_GENERAL_NAME = "general.name";
static const char * LLM_KV_GENERAL_ARCHITECTURE = "general.architecture"; static const char * LLM_KV_GENERAL_ARCHITECTURE = "general.architecture";
static const char * LLM_KV_GENERAL_FILE_TYPE = "general.file_type"; static const char * LLM_KV_GENERAL_FILE_TYPE = "general.file_type";
@ -529,6 +530,7 @@ static void load_llama_model_gguf(struct gguf_context * fctx, struct ggml_contex
static void save_llama_model_gguf(struct gguf_context * fctx, const char * fn_vocab_model, struct my_llama_model * model) { static void save_llama_model_gguf(struct gguf_context * fctx, const char * fn_vocab_model, struct my_llama_model * model) {
const char * arch = "llama"; const char * arch = "llama";
enum llama_ftype ftype = LLAMA_FTYPE_ALL_F32; enum llama_ftype ftype = LLAMA_FTYPE_ALL_F32;
std::vector<char> keybuf; std::vector<char> keybuf;
@ -540,6 +542,7 @@ static void save_llama_model_gguf(struct gguf_context * fctx, const char * fn_vo
// set arch // set arch
gguf_set_val_str(fctx, LLM_KV_GENERAL_ARCHITECTURE, arch); gguf_set_val_str(fctx, LLM_KV_GENERAL_ARCHITECTURE, arch);
gguf_set_val_str(fctx, LLM_KV_GENERAL_NAME, arch);
gguf_set_val_u32(fctx, LLM_KV_GENERAL_FILE_TYPE, ftype); gguf_set_val_u32(fctx, LLM_KV_GENERAL_FILE_TYPE, ftype);
// set hparams // set hparams