tacit/nanochat
1
0
Fork 0
mirror of https://github.com/karpathy/nanochat.git synced 2026-02-22 19:40:25 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Merge 9de0a121c5 into 50413d2d67

Browse Source
This commit is contained in:
Yury Kirpichev 2026-01-16 17:29:48 +10:00 committed by GitHub
commit cb34cc57f1
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 6 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
dev/runcpu.sh
View File

@ -41,7 +41,8 @@ python -m scripts.base_train \
--core-metric-every=50 \
--core-metric-max-per-task=12 \
--sample-every=50 \
--num-iterations=50
--num-iterations=50 \
--run=$WANDB_RUN
python -m scripts.base_loss --device-batch-size=1 --split-tokens=4096
python -m scripts.base_eval --max-per-task=16
@ -52,7 +53,8 @@ python -m scripts.mid_train \
--eval-every=50 \
--eval-tokens=4096 \
--total-batch-size=1024 \
--num-iterations=100
--num-iterations=100 \
--run=$WANDB_RUN
# eval results will be terrible, this is just to execute the code paths.
# note that we lower the execution memory limit to 1MB to avoid warnings on smaller systems
python -m scripts.chat_eval --source=mid --max-new-tokens=128 --max-problems=20
@ -63,7 +65,8 @@ python -m scripts.chat_sft \
--target-examples-per-step=4 \
--num-iterations=100 \
--eval-steps=4 \
--eval-metrics-max-problems=16
--eval-metrics-max-problems=16 \
--run=$WANDB_RUN
# Chat CLI
# python -m scripts.chat_cli -p "Why is the sky blue?"