Files
jspace/scripts/jobs/zzz_q_rungb_lt.sh
T

14 lines
758 B
Bash

# gpuq-in: results-loop/star_data.json results-loop/gsm_cot_data.json results-loop/adapter_carrycot_e400.pt
# gpuq-out: results-loop/eval_gsm_carrycot_b1_lt*.json results-loop/train_carrycot_b1_lt*_log.json results-loop/adapter_carrycot_b1_lt*.pt
git pull origin main -q 2>/dev/null
P=/home/nils/jspace/.venv/bin/python
export JLENS_MODEL=google/gemma-4-E2B-it LOOP_OUT=/home/nils/jspace/results-loop
cd /home/nils/jspace/scripts
for LAM in 0.3 1.0; do
T=lt$(echo $LAM | tr -d .)
$P train_carry_cot.py --drop-steps 1 --warm-start $LOOP_OUT/adapter_carrycot_e400.pt \
--steps 200 --lr 3e-4 --lensteach $LAM
$P eval_carry_cot.py --adapter $LOOP_OUT/adapter_carrycot_b1_${T}_e200.pt \
--tag gsm_carrycot_b1_$T --grid 0:0,2:12,2:16 --n 256
done