mirror of
https://github.com/LostRuins/koboldcpp.git
synced 2026-09-18 16:55:14 +02:00
some tweaks and cleanup
This commit is contained in:
+1
-1
@@ -11124,7 +11124,7 @@ static int llama_decode_impl(
|
||||
|
||||
GGML_ASSERT_CONTINUE(n_tokens_all <= cparams.n_batch);
|
||||
|
||||
GGML_ASSERT_CONTINUE((cparams.causal_attn || cparams.n_ubatch >= n_tokens_all) && "non-causal attention requires n_ubatch >= n_tokens");
|
||||
//GGML_ASSERT_CONTINUE((cparams.causal_attn || cparams.n_ubatch >= n_tokens_all) && "non-causal attention requires n_ubatch >= n_tokens");
|
||||
|
||||
if (lctx.t_compute_start_us == 0) {
|
||||
lctx.t_compute_start_us = ggml_time_us();
|
||||
|
||||
Reference in New Issue
Block a user