Searched defs:q_heads (Results 1 – 1 of 1) sorted by relevance
| /external/pytorch/torch/_inductor/kernel/ |
| D | flex_attention.py | 59 def flex_attention_grid(batch_size, q_heads, num_queries, d_model, meta): argument
910 batch_size, q_heads, num_queries, d_model, kv_heads, num_key_value, meta argument
|