tacit/nanochat
1
0
Fork 0
mirror of https://github.com/karpathy/nanochat.git synced 2026-03-18 02:43:13 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Fix bug in setting precision (#538)

Browse Source
This commit is contained in:
George Shakan 2026-02-18 10:42:11 -05:00 committed by Andrej Karpathy
parent cac43e8511
commit ad55575326
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
nanochat/common.py
View File

@ -170,7 +170,7 @@ def compute_init(device_type="cuda"): # cuda|cpu|mps
# Precision
if device_type == "cuda":
torch.backends.fp32_precision = "tf32" # uses tf32 instead of fp32 for matmuls
torch.set_float32_matmul_precision("high") # uses tf32 instead of fp32 for matmuls, see https://docs.pytorch.org/docs/stable/generated/torch.set_float32_matmul_precision.html
# Distributed setup: Distributed Data Parallel (DDP), optional, and requires CUDA
is_ddp_requested, ddp_rank, ddp_local_rank, ddp_world_size = get_dist_info()