tacit/nanochat
1
0
Fork 0
mirror of https://github.com/karpathy/nanochat.git synced 2026-05-12 19:00:14 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

new optimal ratio for d26 training

Browse Source
This commit is contained in:
Andrej Karpathy 2026-02-06 19:21:27 +00:00
parent e527521a3f
commit 685271dc8d
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
runs/speedrun.sh
View File

@ -70,7 +70,7 @@ echo "Waiting for dataset download to complete..."
wait $DATASET_DOWNLOAD_PID
# d24 model (slightly overtrained is enough to beat GPT-2 => increase data:params ratio from compute optimal 10.5 (default) to 12)
torchrun --standalone --nproc_per_node=8 -m scripts.base_train -- --depth=26 --target-param-data-ratio=8.5 --device-batch-size=16 --fp8 --run=$WANDB_RUN
torchrun --standalone --nproc_per_node=8 -m scripts.base_train -- --depth=26 --target-param-data-ratio=8.25 --device-batch-size=16 --fp8 --run=$WANDB_RUN
# evaluate the model: CORE metric, BPB on train/val, and draw samples
torchrun --standalone --nproc_per_node=8 -m scripts.base_eval -- --device-batch-size=16