upstream-archive/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2026-07-23 11:10:55 -05:00
Code Issues Packages Projects Releases Wiki Activity

model-conversion : cast logits to float32 (#18009)

Browse Source
This commit is contained in:
Georgi Gerganov
2025-12-14 08:58:13 +02:00
committed by GitHub
parent 609a2d0268
commit 77ad8542bd
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
examples/model-conversion/scripts/causal/run-org-model.py
View File

@@ -200,7 +200,7 @@ with torch.no_grad():
logits = outputs.logits
# Extract logits for the last token (next token prediction)
last_logits = logits[0, -1, :].cpu().numpy()
last_logits = logits[0, -1, :].float().cpu().numpy()
print(f"Logits shape: {logits.shape}")
print(f"Last token logits shape: {last_logits.shape}")