File size: 1,118 Bytes
2e30a77 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 | import torch.nn as nn
from model.attention import CausalSelfAttention
from model.mlp import FeedForward
class TransformerBlock(nn.Module):
def __init__(
self,
d_model,
n_heads,
context_length,
dropout=0.1
):
super().__init__()
self.norm1 = nn.LayerNorm(
d_model
)
self.attention = (
CausalSelfAttention(
d_model,
n_heads,
context_length,
dropout
)
)
self.norm2 = nn.LayerNorm(
d_model
)
self.feed_forward = (
FeedForward(
d_model,
dropout
)
)
def forward(self, x, kv_cache=None, cache_position=None):
attn_out = self.attention(
self.norm1(x),
kv_cache=kv_cache,
cache_position=cache_position
)
x = x + attn_out
x = x + self.feed_forward(
self.norm2(x)
)
return x |