mirror of
https://github.com/ollama/ollama.git
synced 2025-05-11 10:26:53 +02:00
llama4
This commit is contained in:
parent
54055a6dae
commit
f0c66e6dea
13 changed files with 833 additions and 15 deletions
|
@ -133,6 +133,7 @@ type Tensor interface {
|
|||
Mul(ctx Context, t2 Tensor) Tensor
|
||||
Mulmat(ctx Context, t2 Tensor) Tensor
|
||||
MulmatFullPrec(ctx Context, t2 Tensor) Tensor
|
||||
MulmatID(ctx Context, t2, ids Tensor) Tensor
|
||||
|
||||
Softmax(ctx Context) Tensor
|
||||
LayerNorm(ctx Context, weight, bias Tensor, eps float32) Tensor
|
||||
|
@ -150,6 +151,7 @@ type Tensor interface {
|
|||
Tanh(ctx Context) Tensor
|
||||
GELU(ctx Context) Tensor
|
||||
SILU(ctx Context) Tensor
|
||||
Sigmoid(ctx Context) Tensor
|
||||
|
||||
Reshape(ctx Context, shape ...int) Tensor
|
||||
View(ctx Context, offset int, shape ...int) Tensor
|
||||
|
@ -168,6 +170,8 @@ type Tensor interface {
|
|||
Rows(ctx Context, t2 Tensor) Tensor
|
||||
Copy(ctx Context, t2 Tensor) Tensor
|
||||
Duplicate(ctx Context) Tensor
|
||||
|
||||
TopK(ctx Context, k int) Tensor
|
||||
}
|
||||
|
||||
// ScaledDotProductAttention implements a fused attention
|
||||
|
|
Loading…
Add table
Add a link
Reference in a new issue