Avoid the attention mask where possible. (#1933)

1 files changed, 20 insertions, 12 deletions

diff --git a/candle-transformers/src/models/falcon.rs b/candle-transformers/src/models/falcon.rs
index 86cf8451..24fd3c46 100644
--- a/candle-transformers/src/models/falcon.rs
+++ b/candle-transformers/src/models/falcon.rs
@@ -247,7 +247,7 @@ impl FalconAttention {
         }
     }
 
-    fn forward(&mut self, x: &Tensor, mask: &Tensor, past_kv_len: usize) -> Result<Tensor> {
+    fn forward(&mut self, x: &Tensor, mask: Option<&Tensor>, past_kv_len: usize) -> Result<Tensor> {
         let fused_qkv = self.query_key_value.forward(x)?;
         let head_dim = self.head_dim;
         let (query, key, value) = self.split_heads(&fused_qkv)?;
@@ -267,7 +267,6 @@ impl FalconAttention {
             (query, key)
         };
         let (mut key, mut value) = (key, value);
-        let mask = masked_fill(&mask.to_dtype(DType::F32)?, mask, -1e9)?.to_dtype(query.dtype())?;
         if self.use_cache {
             if let Some((cache_k, cache_v)) = &self.kv_cache {
                 // TODO: we could trim the tensors to MAX_SEQ_LEN so that this would work for
@@ -293,13 +292,18 @@ impl FalconAttention {
 
         // Only handle the case where alibi is None here, and non-flash attention.
         let attention_scores = (query.matmul(&key.t()?)? * self.inv_norm_factor)?;
-        let attention_scores = candle_nn::ops::softmax(
-            &attention_scores
-                .broadcast_add(&mask.squeeze(1)?)?
-                .to_dtype(DType::F32)?,
-            D::Minus1,
-        )?
-        .to_dtype(x.dtype())?;
+        let attention_scores = match mask {
+            None => attention_scores,
+            Some(mask) => {
+                let mask = masked_fill(&mask.to_dtype(DType::F32)?, mask, -1e9)?
+                    .to_dtype(query.dtype())?;
+                attention_scores.broadcast_add(&mask.squeeze(1)?)?
+            }
+        };
+
+        let attention_scores =
+            candle_nn::ops::softmax(&attention_scores.to_dtype(DType::F32)?, D::Minus1)?
+                .to_dtype(x.dtype())?;
         let attn_output = attention_scores
             .matmul(&value)?
             .reshape((b_sz, self.num_heads, seq_len, head_dim))?
@@ -372,7 +376,7 @@ impl FalconDecoderLayer {
         })
     }
 
-    fn forward(&mut self, x: &Tensor, mask: &Tensor, past_kv_len: usize) -> Result<Tensor> {
+    fn forward(&mut self, x: &Tensor, mask: Option<&Tensor>, past_kv_len: usize) -> Result<Tensor> {
         let residual = x.clone();
         let ln_attn = self.inp_layernorm.forward(x)?;
         let attn_output = self.self_attention.forward(&ln_attn, mask, past_kv_len)?;
@@ -457,9 +461,13 @@ impl Falcon {
             Some((k, _)) => k.dim(1)?,
             None => 0,
         };
-        let causal_mask = prepare_attn_mask(b_sz, seq_len)?.to_device(input_ids.device())?;
+        let causal_mask = if seq_len <= 1 {
+            None
+        } else {
+            Some(prepare_attn_mask(b_sz, seq_len)?.to_device(input_ids.device())?)
+        };
         for block in self.blocks.iter_mut() {
-            hidden_state = block.forward(&hidden_state, &causal_mask, past_kv_len)?;
+            hidden_state = block.forward(&hidden_state, causal_mask.as_ref(), past_kv_len)?;
         }
         let hidden_state = self.ln_f.forward(&hidden_state)?;
         let hidden_state = hidden_state.narrow(1, seq_len - 1, 1)?;