mirror of
https://github.com/ml-explore/mlx.git
synced 2025-12-08 12:18:57 +08:00
more accurate rope fallback (#2792)
This commit is contained in:
20
mlx/fast.cpp
20
mlx/fast.cpp
@@ -416,23 +416,25 @@ array rope(
|
||||
if (offset.size() > 1) {
|
||||
offset = expand_dims(offset, {-1, -2}, s);
|
||||
}
|
||||
auto positions =
|
||||
multiply(add(arange(x.shape(2), t, s), offset, s), array(scale, t), s);
|
||||
auto positions = multiply(
|
||||
add(arange(x.shape(2), float32, s), offset, s),
|
||||
array(scale, float32),
|
||||
s);
|
||||
|
||||
auto default_inv_freqs = [&s, &t, base, half_dims]() {
|
||||
auto default_inv_freqs = [&s, base, half_dims]() {
|
||||
return exp(
|
||||
multiply(
|
||||
arange(0, -half_dims, -1, t, s),
|
||||
array(std::log(base) / half_dims, t),
|
||||
arange(0, -half_dims, -1, float32, s),
|
||||
array(std::log(base) / half_dims, float32),
|
||||
s),
|
||||
s);
|
||||
};
|
||||
|
||||
auto inv_freqs = inputs.size() == 3 ? astype(reciprocal(inputs[2], s), t, s)
|
||||
: default_inv_freqs();
|
||||
auto inv_freqs =
|
||||
inputs.size() == 3 ? reciprocal(inputs[2], s) : default_inv_freqs();
|
||||
auto theta = multiply(expand_dims(positions, -1, s), inv_freqs, s);
|
||||
auto coss = cos(theta, s);
|
||||
auto sins = sin(theta, s);
|
||||
auto coss = astype(cos(theta, s), t, s);
|
||||
auto sins = astype(sin(theta, s), t, s);
|
||||
|
||||
auto apply_rope = [forward, s](
|
||||
const array& x1,
|
||||
|
||||
@@ -332,6 +332,26 @@ class TestFast(mlx_tests.MLXTestCase):
|
||||
rx = rope_orig(x, dims, traditional, base, scale, offset)
|
||||
self.assertLess(mx.abs(rx - rx_fast).max(), 1e-5)
|
||||
|
||||
def test_rope_with_large_offset(self):
|
||||
x = mx.random.normal(shape=(1, 1, 1024, 32))
|
||||
rx_fp32 = mx.fast.rope(
|
||||
x,
|
||||
32,
|
||||
traditional=False,
|
||||
scale=1.0,
|
||||
base=10000,
|
||||
offset=4000,
|
||||
)
|
||||
rx_bf16 = mx.fast.rope(
|
||||
x.astype(mx.bfloat16),
|
||||
32,
|
||||
traditional=False,
|
||||
scale=1.0,
|
||||
base=10000,
|
||||
offset=4000,
|
||||
)
|
||||
self.assertLess((rx_fp32 - rx_bf16).abs().max(), 1e-1)
|
||||
|
||||
def test_rms_norm(self):
|
||||
# Per dtype absolute tolerance
|
||||
tolerances = {mx.float32: 1e-6, mx.float16: 1e-3, mx.bfloat16: 1e-2}
|
||||
|
||||
Reference in New Issue
Block a user