Merge branch 'main' into swa_left_brcm_padding

Signed-off-by: Charlene Yang <[email protected]>

tests/jax/conftest.py

@@ -20,7 +20,7 @@ def clear_live_arrays():
 
 
 @pytest.fixture(autouse=True, scope="module")
-def enable_fused_attn():
+def enable_fused_attn_after_hopper():
     """
     Enable fused attn for hopper+ arch.
     Fused attn kernels on pre-hopper arch are not deterministic.

tests/jax/test_distributed_fused_attn.py

@@ -20,7 +20,6 @@
 from utils import (
     make_causal_mask,
     make_self_mask,
-    assert_tree_like_allclose,
     assert_allclose,
     print_debug_tensor_stats,
 )
@@ -32,7 +31,6 @@
     AttnMaskType,
     QKVLayout,
     QKVFormat,
-    get_qkv_format,
     reorder_causal_load_balancing,
     inverse_reorder_causal_load_balancing,
     CPStrategy,
@@ -421,7 +419,7 @@ def impl_test_contex_parallel_attn(
         dropout_prob = 0.0
         is_training = True
         dp_size, cp_size, tp_size = mesh_shape
-        qkv_format = get_qkv_format(qkv_layout)
+        qkv_format = qkv_layout.get_qkv_format()
 
         batch, seqlen, num_head, hidden = data_shape
 
@@ -503,7 +501,7 @@ def grad_func(func, *args, **kwargs):
             # Gradient is small, use a gradient multiplier to amplify the gradient
             _, max_seq_len, num_heads, _ = data_shape
             gradient_multiplier = max_seq_len * num_heads
-            if attn_mask_type in [AttnMaskType.CAUSAL_MASK, AttnMaskType.CAUSAL_BOTTOM_RIGHT_MASK]:
+            if attn_mask_type.is_causal():
                 gradient_multiplier /= 10
             ret_valid = func(*args, **kwargs)
             return (jnp.mean(ret_valid, dtype=jnp.float32) * gradient_multiplier).astype(dtype)