Rename to llm_build_ffn_mpt_awq
This commit is contained in:
Le Hoang Anh
parent
d2e9d00cbc
commit
0610672b19
3 changed files with 36 additions and 68 deletions
|
|
@ -1100,11 +1100,13 @@ void llama_batch_add(
|
|||
|
||||
std::tuple<struct llama_model *, struct llama_context *> llama_init_from_gpt_params(gpt_params & params) {
|
||||
auto mparams = llama_model_params_from_gpt_params(params);
|
||||
|
||||
llama_model * model = llama_load_model_from_file(params.model.c_str(), mparams);
|
||||
if (model == NULL) {
|
||||
fprintf(stderr, "%s: error: failed to load model '%s'\n", __func__, params.model.c_str());
|
||||
return std::make_tuple(nullptr, nullptr);
|
||||
}
|
||||
|
||||
auto cparams = llama_context_params_from_gpt_params(params);
|
||||
|
||||
llama_context * lctx = llama_new_context_with_model(model, cparams);
|
||||
|
|
|
|||
72
llama.cpp
72
llama.cpp
|
|
@ -454,8 +454,8 @@ static std::map<llm_arch, std::map<llm_tensor, std::string>> LLM_TENSOR_NAMES =
|
|||
{ LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
|
||||
{ LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
|
||||
{ LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
|
||||
{ LLM_TENSOR_FFN_ACT, "blk.%d.ffn.act"},
|
||||
{ LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
|
||||
{ LLM_TENSOR_FFN_ACT, "blk.%d.ffn.act" },
|
||||
},
|
||||
},
|
||||
{
|
||||
|
|
@ -1178,6 +1178,7 @@ struct llama_hparams {
|
|||
|
||||
float f_clamp_kqv;
|
||||
float f_max_alibi_bias;
|
||||
|
||||
bool use_awq;
|
||||
|
||||
bool operator!=(const llama_hparams & other) const {
|
||||
|
|
@ -3423,10 +3424,10 @@ static void llm_load_tensors(
|
|||
layer.ffn_norm = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_NORM, "weight", i), {n_embd}, backend);
|
||||
|
||||
layer.ffn_down = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_DOWN, "weight", i), { n_ff, n_embd}, backend_split);
|
||||
layer.ffn_up = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_UP, "weight", i), {n_embd, n_ff}, backend_split);
|
||||
if (model.hparams.use_awq) {
|
||||
layer.ffn_act = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_ACT, "scales", i), {n_ff}, backend);
|
||||
}
|
||||
layer.ffn_up = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_UP, "weight", i), {n_embd, n_ff}, backend_split);
|
||||
|
||||
if (backend == GGML_BACKEND_GPU) {
|
||||
if (model.hparams.use_awq) {
|
||||
|
|
@ -3436,10 +3437,9 @@ static void llm_load_tensors(
|
|||
ggml_nbytes(layer.wo) +
|
||||
ggml_nbytes(layer.ffn_norm) +
|
||||
ggml_nbytes(layer.ffn_down) +
|
||||
ggml_nbytes(layer.ffn_act) +
|
||||
ggml_nbytes(layer.ffn_up);
|
||||
}
|
||||
else {
|
||||
ggml_nbytes(layer.ffn_up) +
|
||||
ggml_nbytes(layer.ffn_act);
|
||||
} else {
|
||||
vram_weights +=
|
||||
ggml_nbytes(layer.attn_norm) +
|
||||
ggml_nbytes(layer.wqkv) +
|
||||
|
|
@ -3648,6 +3648,7 @@ static bool llama_model_load(const std::string & fname, llama_model & model, con
|
|||
|
||||
model.hparams.vocab_only = params.vocab_only;
|
||||
model.hparams.use_awq = params.use_awq;
|
||||
|
||||
llm_load_arch (ml, model);
|
||||
llm_load_hparams(ml, model);
|
||||
llm_load_vocab (ml, model);
|
||||
|
|
@ -3935,7 +3936,7 @@ static struct ggml_tensor * llm_build_ffn(
|
|||
return cur;
|
||||
}
|
||||
|
||||
static struct ggml_tensor *llm_build_ffn(
|
||||
static struct ggml_tensor * llm_build_ffn_mpt_awq(
|
||||
struct ggml_context *ctx,
|
||||
struct ggml_tensor *cur,
|
||||
struct ggml_tensor *up,
|
||||
|
|
@ -3953,43 +3954,14 @@ static struct ggml_tensor *llm_build_ffn(
|
|||
struct ggml_tensor * tmp = ggml_mul_mat(ctx, up, cur);
|
||||
cb(tmp, "ffn_up", il);
|
||||
|
||||
if (up_b)
|
||||
{
|
||||
if (up_b) {
|
||||
tmp = ggml_add(ctx, tmp, up_b);
|
||||
cb(tmp, "ffn_up_b", il);
|
||||
}
|
||||
|
||||
if (gate)
|
||||
{
|
||||
switch (type_gate)
|
||||
{
|
||||
case LLM_FFN_SEQ:
|
||||
{
|
||||
cur = ggml_mul_mat(ctx, gate, tmp);
|
||||
cb(cur, "ffn_gate", il);
|
||||
}
|
||||
break;
|
||||
case LLM_FFN_PAR:
|
||||
{
|
||||
cur = ggml_mul_mat(ctx, gate, cur);
|
||||
cb(cur, "ffn_gate", il);
|
||||
}
|
||||
break;
|
||||
}
|
||||
|
||||
if (gate_b)
|
||||
{
|
||||
cur = ggml_add(ctx, cur, gate_b);
|
||||
cb(cur, "ffn_gate_b", il);
|
||||
}
|
||||
}
|
||||
else
|
||||
{
|
||||
cur = tmp;
|
||||
}
|
||||
|
||||
switch (type_op)
|
||||
{
|
||||
switch (type_op) {
|
||||
case LLM_FFN_GELU_ACT:
|
||||
{
|
||||
cur = ggml_gelu(ctx, cur);
|
||||
|
|
@ -3998,24 +3970,20 @@ static struct ggml_tensor *llm_build_ffn(
|
|||
cb(repeat, "ffn_repeat(scales)", il);
|
||||
cur = ggml_div(ctx, cur, repeat);
|
||||
cb(cur, "ffn_div(gelu)", il);
|
||||
}
|
||||
break;
|
||||
} break;
|
||||
}
|
||||
|
||||
if (type_gate == LLM_FFN_PAR)
|
||||
{
|
||||
if (type_gate == LLM_FFN_PAR) {
|
||||
cur = ggml_mul(ctx, cur, tmp);
|
||||
cb(cur, "ffn_gate_par", il);
|
||||
}
|
||||
|
||||
cur = ggml_mul_mat(ctx, down, cur);
|
||||
if (down_b)
|
||||
{
|
||||
if (down_b) {
|
||||
cb(cur, "ffn_down", il);
|
||||
}
|
||||
|
||||
if (down_b)
|
||||
{
|
||||
if (down_b) {
|
||||
cur = ggml_add(ctx, cur, down_b);
|
||||
}
|
||||
|
||||
|
|
@ -5133,21 +5101,17 @@ struct llm_build_context {
|
|||
LLM_NORM, cb, il);
|
||||
cb(cur, "ffn_norm", il);
|
||||
if (hparams.use_awq) {
|
||||
cur = llm_build_ffn(ctx0, cur,
|
||||
cur = llm_build_ffn_mpt_awq(ctx0, cur,
|
||||
model.layers[il].ffn_up, NULL,
|
||||
NULL, NULL,
|
||||
model.layers[il].ffn_down, NULL,
|
||||
model.layers[il].ffn_act,
|
||||
LLM_FFN_GELU_ACT, LLM_FFN_SEQ, cb, il);
|
||||
|
||||
}
|
||||
else {
|
||||
} else {
|
||||
cur = llm_build_ffn(ctx0, cur,
|
||||
model.layers[il].ffn_up, NULL,
|
||||
NULL, NULL,
|
||||
model.layers[il].ffn_down, NULL,
|
||||
LLM_FFN_GELU, LLM_FFN_SEQ, cb, il);
|
||||
|
||||
}
|
||||
cb(cur, "ffn_out", il);
|
||||
}
|
||||
|
|
@ -8864,9 +8828,9 @@ struct llama_model_params llama_model_default_params() {
|
|||
/*.progress_callback_user_data =*/ nullptr,
|
||||
/*.kv_overrides =*/ nullptr,
|
||||
/*.vocab_only =*/ false,
|
||||
/*.use_awq =*/ false,
|
||||
/*.use_mmap =*/ true,
|
||||
/*.use_mlock =*/ false,
|
||||
/*.use_awq =*/ false,
|
||||
};
|
||||
|
||||
#ifdef GGML_USE_METAL
|
||||
|
|
@ -8960,7 +8924,9 @@ struct llama_model * llama_load_model_from_file(
|
|||
const char * path_model,
|
||||
struct llama_model_params params) {
|
||||
ggml_time_init();
|
||||
|
||||
llama_model * model = new llama_model;
|
||||
|
||||
unsigned cur_percentage = 0;
|
||||
if (params.progress_callback == NULL) {
|
||||
params.progress_callback_user_data = &cur_percentage;
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue