]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
Set GLM4 blk.*.attn_output.weight, kqv_out-* matmul to GGML_PREC_F32 to fix infinity...
author0cc4m <redacted>
Tue, 20 May 2025 08:11:56 +0000 (10:11 +0200)
committerGitHub <redacted>
Tue, 20 May 2025 08:11:56 +0000 (10:11 +0200)
commitc9c64dee572c5eedf073a6c6eb5d92d8283f7639
tree92b77fb63297d392bacd3f3b7da59de9964ba266
parentc00a2634bec49805c6b31438b1a6006a2bd793cb
Set GLM4 blk.*.attn_output.weight, kqv_out-* matmul to GGML_PREC_F32 to fix infinity values in output (#13639)
src/llama-graph.cpp