3×10×256
3×10×256
var_0
float32[3,10,256]
deterministic
boolean
TransformerStack_1
TransformerStack_1
ƒ
Show Function Definition
var_418
float32[3,10,256]
×
transformer_stack
TransformerStack_1
TransformerBlock_2
MultiHeadAttention_2
attention_0
FeedForward_2
TransformerBlock_5
MultiHeadAttention_5
attention_1
FeedForward_5
TransformerBlock_8
MultiHeadAttention_8
attention_2
FeedForward_8
TransformerBlock_11
MultiHeadAttention_11
attention_3
FeedForward_11
TransformerBlock_14
MultiHeadAttention_14
attention_4
FeedForward_14
TransformerBlock_17
MultiHeadAttention_17
attention_5
FeedForward_17
❮
Version
{version}
Copyright ©
Lutz Roeder
Open Model…
.
.
.
OK
≡