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

Print model version.

Browse Source 
Also improve model type printing, and fix indentation of an unrelated
switch statement.
This commit is contained in:
comex 2023-04-08 13:08:21 -07:00 committed by Matvey Soloviev
parent f963b63afa
commit 180b693a47
1 changed files with 46 additions and 23 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

69
llama.cpp
View File

@ -356,17 +356,17 @@ struct llama_load_tensor {
LLAMA_ASSERT(shards.size() <= UINT32_MAX); LLAMA_ASSERT(shards.size() <= UINT32_MAX);
uint32_t n_shards = (uint32_t) shards.size(); uint32_t n_shards = (uint32_t) shards.size();
switch (split_type) { switch (split_type) {
case SPLIT_NONE: case SPLIT_NONE:
ne = first_shard.ne; ne = first_shard.ne;
break; break;
case SPLIT_BY_COLUMNS: case SPLIT_BY_COLUMNS:
ne = {checked_mul<uint32_t>(first_shard.ne[0], n_shards), ne = {checked_mul<uint32_t>(first_shard.ne[0], n_shards),
first_shard.ne[1]}; first_shard.ne[1]};
break; break;
case SPLIT_BY_ROWS: case SPLIT_BY_ROWS:
ne = {first_shard.ne[0], ne = {first_shard.ne[0],
checked_mul<uint32_t>(first_shard.ne[1], n_shards)}; checked_mul<uint32_t>(first_shard.ne[1], n_shards)};
break; break;
} }
} }
@ -806,6 +806,25 @@ bool llama_mlock_supported() {
// model loading // model loading
// //
static const char *llama_file_version_name(llama_file_version version) {
switch (version) {
case LLAMA_FILE_VERSION_GGML: return "'ggml' (old version with low tokenizer quality and no mmap support)";
case LLAMA_FILE_VERSION_GGMF_V1: return "ggmf v1 (old version with no mmap support)";
case LLAMA_FILE_VERSION_GGJT_V1: return "ggjt v1 (latest)";
default: LLAMA_ASSERT(false);
}
}
static const char *llama_model_type_name(e_model type) {
switch (type) {
case MODEL_7B: return "7B";
case MODEL_13B: return "13B";
case MODEL_30B: return "30B";
case MODEL_65B: return "65B";
default: LLAMA_ASSERT(false);
}
}
static void llama_model_load_internal( static void llama_model_load_internal(
const std::string & fname, const std::string & fname,
llama_context & lctx, llama_context & lctx,
@ -823,8 +842,9 @@ static void llama_model_load_internal(
lctx.vocab = std::move(ml->file_loaders.at(0)->vocab); lctx.vocab = std::move(ml->file_loaders.at(0)->vocab);
auto & model = lctx.model; auto & model = lctx.model;
model.hparams = ml->file_loaders.at(0)->hparams;
llama_file_version file_version = ml->file_loaders.at(0)->file_version;
auto & hparams = model.hparams; auto & hparams = model.hparams;
hparams = ml->file_loaders.at(0)->hparams;
uint32_t n_ff = ((2*(4*hparams.n_embd)/3 + hparams.n_mult - 1)/hparams.n_mult)*hparams.n_mult; uint32_t n_ff = ((2*(4*hparams.n_embd)/3 + hparams.n_mult - 1)/hparams.n_mult)*hparams.n_mult;
{ {
@ -836,18 +856,21 @@ static void llama_model_load_internal(
} }
hparams.n_ctx = n_ctx; hparams.n_ctx = n_ctx;
}
fprintf(stderr, "%s: n_vocab = %u\n", __func__, hparams.n_vocab); {
fprintf(stderr, "%s: n_ctx = %u\n", __func__, hparams.n_ctx); fprintf(stderr, "%s: format = %s\n", __func__, llama_file_version_name(file_version));
fprintf(stderr, "%s: n_embd = %u\n", __func__, hparams.n_embd); fprintf(stderr, "%s: n_vocab = %u\n", __func__, hparams.n_vocab);
fprintf(stderr, "%s: n_mult = %u\n", __func__, hparams.n_mult); fprintf(stderr, "%s: n_ctx = %u\n", __func__, hparams.n_ctx);
fprintf(stderr, "%s: n_head = %u\n", __func__, hparams.n_head); fprintf(stderr, "%s: n_embd = %u\n", __func__, hparams.n_embd);
fprintf(stderr, "%s: n_layer = %u\n", __func__, hparams.n_layer); fprintf(stderr, "%s: n_mult = %u\n", __func__, hparams.n_mult);
fprintf(stderr, "%s: n_rot = %u\n", __func__, hparams.n_rot); fprintf(stderr, "%s: n_head = %u\n", __func__, hparams.n_head);
fprintf(stderr, "%s: f16 = %u\n", __func__, hparams.f16); fprintf(stderr, "%s: n_layer = %u\n", __func__, hparams.n_layer);
fprintf(stderr, "%s: n_ff = %u\n", __func__, n_ff); fprintf(stderr, "%s: n_rot = %u\n", __func__, hparams.n_rot);
fprintf(stderr, "%s: n_parts = %zu\n", __func__, ml->file_loaders.size()); fprintf(stderr, "%s: f16 = %u\n", __func__, hparams.f16);
fprintf(stderr, "%s: type = %u\n", __func__, model.type); fprintf(stderr, "%s: n_ff = %u\n", __func__, n_ff);
fprintf(stderr, "%s: n_parts = %zu\n", __func__, ml->file_loaders.size());
fprintf(stderr, "%s: model size = %s\n", __func__, llama_model_type_name(model.type));
} }
if (vocab_only) { if (vocab_only) {