minor : fix GPT-NeoX name

diff --git a/README.md b/README.md
index ac8240e0035bf107988892c0dcc3e0f8dee32376..c69d790d6715ed684dc5934ec07f4a528679606c 100644 (file)
--- a/README.md
+++ b/README.md
@@ -29,7 +29,7 @@ Some of the development is currently happening in the [llama.cpp](https://github
 - [X] Example of RWKV inference [saharNooby/rwkv.cpp](https://github.com/saharNooby/rwkv.cpp)
 - [ ] Example of [SAM](https://github.com/facebookresearch/segment-anything) inference
 - [ ] Idea for GPU support: https://github.com/ggerganov/llama.cpp/discussions/915
-- [X] Example of StableLM (GPTNeoX) inference [examples/stablelm](https://github.com/ggerganov/ggml/tree/master/examples/stablelm)
+- [X] Example of StableLM (GPT-NeoX) inference [examples/stablelm](https://github.com/ggerganov/ggml/tree/master/examples/stablelm)
 
 ## Whisper inference (example)
 

diff --git a/examples/stablelm/README.md b/examples/stablelm/README.md
index 81d18dcf0afc4b185e1e86a983b047d6c081c96e..7882d8b34957a983400cd7c8983964afd3d9088f 100644 (file)
--- a/examples/stablelm/README.md
+++ b/examples/stablelm/README.md
@@ -1,6 +1,6 @@
 # StableLM
 
-Transformer architecture: GPTNeoX
+Transformer architecture: GPT-NeoX
 
 Ref: https://github.com/stability-AI/stableLM/#stablelm-alpha
 

diff --git a/examples/stablelm/main.cpp b/examples/stablelm/main.cpp
index a3c10ed455f964af48b61ed26fd4895ee947aad1..7cccb687a47d090ae9530b9399c9614c10cfb13b 100644 (file)
--- a/examples/stablelm/main.cpp
+++ b/examples/stablelm/main.cpp
@@ -480,7 +480,7 @@ bool stablelm_eval(
             struct ggml_tensor * Kcur = ggml_cont(ctx0, ggml_view_3d(ctx0, cur, n_embd/n_head, n_head, N, cur->nb[1]/n_head, cur->nb[1], 1*sizeof(float)*n_embd/n_head));
             struct ggml_tensor * Vcur = ggml_cont(ctx0, ggml_view_3d(ctx0, cur, n_embd/n_head, n_head, N, cur->nb[1]/n_head, cur->nb[1], 2*sizeof(float)*n_embd/n_head));
 
-            // using mode = 2 for GPTNeoX mode
+            // using mode = 2 for GPT-NeoX mode
             Qcur = ggml_rope(ctx0, Qcur, n_past, n_rot, 2);
             Kcur = ggml_rope(ctx0, Kcur, n_past, n_rot, 2);
 

diff --git a/include/ggml/ggml.h b/include/ggml/ggml.h
index 51cab01fdad3730559da593dd730944efd21f327..460d4ffe03d85adb427cb916369186edb6ba3b54 100644 (file)
--- a/include/ggml/ggml.h
+++ b/include/ggml/ggml.h
@@ -631,7 +631,7 @@ struct ggml_tensor * ggml_soft_max(
 // rotary position embedding
 // in-place, returns view(a)
 // if mode & 1 == 1, skip n_past elements
-// if mode & 2 == 1, GPTNeoX style
+// if mode & 2 == 1, GPT-NeoX style
 // TODO: avoid creating a new tensor every time
 struct ggml_tensor * ggml_rope(
         struct ggml_context * ctx,