vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

examples : fix gpt-neox (#2943)

Browse source 
Co-authored-by: mmnga <mmnga1mmnga@gmail.com>
This commit is contained in:
momonga 2023-09-03 14:36:28 +09:00 • committed by GitHub
parent 2753415afd
commit c42f0ec6b3
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
2 changed files with 51 additions and 8 deletions
Download patch file Download diff file Expand all files Collapse all files

13
examples/gptneox-wip/gptneox-main.cpp
View file

@ -660,9 +660,10 @@ bool gpt_neox_model_load(const std::string & fname, gpt_neox_model & model, gpt2
ggml_tensor * gpt_neox_ff(
const gpt_neox_block &block,
ggml_context * ctx0,
ggml_tensor * inp) {
ggml_tensor * inp,
const gpt_neox_hparams &hparams) {
ggml_tensor * cur = ggml_norm(ctx0, inp);
ggml_tensor * cur = ggml_norm(ctx0, inp, hparams.norm_eps);
cur = ggml_add(ctx0, ggml_mul(ctx0, ggml_repeat(ctx0, block.ln_2_g, cur), cur), ggml_repeat(ctx0, block.ln_2_b, cur));
cur = ggml_mul_mat(ctx0, block.c_mlp_fc_w, cur);
@ -753,7 +754,7 @@ bool gpt_neox_eval(
// self-attention
{
{
cur = ggml_norm(ctx0, inpL);
cur = ggml_norm(ctx0, inpL, hparams.norm_eps);
cur = ggml_add(ctx0,
ggml_mul(ctx0, ggml_repeat(ctx0, model.blocks[il].ln_1_g, cur), cur),
@ -844,7 +845,7 @@ bool gpt_neox_eval(
if (hparams.par_res == 0) {
struct ggml_tensor * inpFF = ggml_add(ctx0, cur, inpL);
cur = gpt_neox_ff(model.blocks[il], ctx0, inpFF);
cur = gpt_neox_ff(model.blocks[il], ctx0, inpFF, hparams);
// input for next layer
inpL = ggml_add(ctx0, cur, inpFF);
@ -853,7 +854,7 @@ bool gpt_neox_eval(
// this is independent of the self-attention result, so it could be done in parallel to the self-attention
// note here we pass inpL instead of cur
cur = gpt_neox_ff(model.blocks[il], ctx0, inpL);
cur = gpt_neox_ff(model.blocks[il], ctx0, inpL, hparams);
// layer input + FF
cur = ggml_add(ctx0, cur, inpFF);
@ -867,7 +868,7 @@ bool gpt_neox_eval(
// norm
{
inpL = ggml_norm(ctx0, inpL);
inpL = ggml_norm(ctx0, inpL, hparams.norm_eps);
// inpL = ln_f_g*inpL + ln_f_b
inpL = ggml_add(ctx0,