tacit/nanochat
1
0
Fork 0
mirror of https://github.com/karpathy/nanochat.git synced 2025-12-15 16:52:14 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

cleaning up speedrun.sh

Browse Source
This commit is contained in:
Sachin Agrawal 2025-11-03 13:10:04 +01:00
parent cf5e213613
commit 4163c648c6
1 changed files with 0 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
speedrun.sh
View File

@ -76,14 +76,6 @@ python -m scripts.tok_eval
# ----------------------------------------------------------------------------- # -----------------------------------------------------------------------------
# Base model (pretraining) # Base model (pretraining)
# Download the eval_bundle from s3 to evaluate CORE metric during training (~162MB)
EVAL_BUNDLE_URL=https://karpathy-public.s3.us-west-2.amazonaws.com/eval_bundle.zip
if [ ! -d "$NANOCHAT_BASE_DIR/eval_bundle" ]; then
curl -L -o eval_bundle.zip $EVAL_BUNDLE_URL
unzip -q eval_bundle.zip
rm eval_bundle.zip
mv eval_bundle $NANOCHAT_BASE_DIR
fi
# The d20 model is 561M parameters. # The d20 model is 561M parameters.
# Chinchilla says #tokens = 20X #params, so we need 561e6 * 20 = 11.2B tokens. # Chinchilla says #tokens = 20X #params, so we need 561e6 * 20 = 11.2B tokens.