[Misc] Standardize RoPE handling for Qwen2-VL (#9250)

2024-10-16 13:56:17 +08:00
parent ed920135c8
commit 7e7eae338d
16 changed files with 102 additions and 200 deletions
--- a/benchmarks/kernels/benchmark_rope.py
+++ b/benchmarks/kernels/benchmark_rope.py
@@ -31,7 +31,7 @@ def benchmark_rope_kernels_multi_lora(
    # batched RoPE can take multiple scaling factors
    batched_rope = get_rope(head_size, rotary_dim, max_position, base,
                            is_neox_style, {
-                                "type": "linear",
+                                "rope_type": "linear",
                                "factor": tuple(scaling_factors)
                            })
    # non-batched RoPE takes only one scaling factor, we create multiple
@@ -41,7 +41,7 @@ def benchmark_rope_kernels_multi_lora(
        non_batched_ropes.append(
            get_rope(head_size, rotary_dim, max_position, base, is_neox_style,
                     {
-                         "type": "linear",
+                         "rope_type": "linear",
                         "factor": (scaling_factor, )
                     }))