code/slurm_logs/mega_extrawide.sbatch

#!/bin/bash
#SBATCH --job-name=vl
#SBATCH --partition=gpu-large
#SBATCH --qos=batch-long
#SBATCH --gres=gpu:1
#SBATCH --cpus-per-task=2
#SBATCH --mem=64G
#SBATCH --time=04:00:00
#SBATCH --output=/weka/s225250685/mats-tist/slurm_logs/mega_d_%j.out

set -u
cd /weka/s225250685/mats-tist

export HF_HOME=/weka/s225250685/Huggingface
export HF_HUB_CACHE=/weka/s225250685/Huggingface/hub
export DB_EXEC_API_DISABLE=1
export PYTHONNOUSERSITE=1
export NO_PROXY=localhost,127.0.0.1
export PYTHONPATH=/weka/s225250685/mats-tist

PY=/weka/s225250685/conda-envs/handbook/bin/python
VLLM=/weka/s225250685/conda-envs/handbook/bin/vllm
PLANNER=/weka/s225250685/mats-tist/alignment-handbook/output/planner-iter2-collab-3B
SEL_V2=/weka/s225250685/mats-tist/alignment-handbook/output/selector-3B-v2-rows

LOG=/weka/s225250685/mats-tist/slurm_logs/mega_d_${SLURM_JOB_ID}.log
: > "$LOG"

nvidia-smi --query-gpu=name,memory.total --format=csv,noheader | tee -a "$LOG"

kill_vllm() {
    pkill -9 -f "vllm serve" 2>/dev/null || true
    pkill -9 -f "VLLM::EngineCore" 2>/dev/null || true
    sleep 5
}
trap kill_vllm EXIT

wait_url() {
    for i in {1..180}; do
        curl --noproxy '*' -fs "$1" >/dev/null 2>&1 && return 0
        sleep 5
    done
    return 1
}

##############################################
# STAGE A: 1-stage K=8 EXTRA-wide mixed-temp (0.2/0.5/0.8/1.1/1.4)
##############################################
kill_vllm
echo "==== [A] launching planner iter-2 vLLM ====" | tee -a "$LOG"
$VLLM serve "$PLANNER" \
    --served-model-name planner --port 8100 --dtype bfloat16 \
    --gpu-memory-utilization 0.85 --enforce-eager --max-model-len 8192 \
    > "${LOG}.serve_p" 2>&1 &
wait_url http://localhost:8100/v1/models && echo "planner READY" | tee -a "$LOG"

OUT_EW=eval_results/scaleup_BoN8_d_K8_1stage_planner_iter2_extrawidetemp_bird_dev.jsonl
rm -f "$OUT_EW"

echo "==== [A] K=8 1-stage EXTRA-wide mixed-temp (0.2,0.5,0.8,1.1,1.4) top_p=0.98 ====" | tee -a "$LOG"
$PY scripts/run_pipeline_rollouts.py \
    --input_file data/sft_bird_with_evidence_dev_text2sql.json \
    --output_file "$OUT_EW" \
    --planner_host http://localhost:8100 \
    --validator_host none \
    --fixer_host none \
    --K 8 --K_val 1 --K_fix 1 \
    --temperature 1.0 --top_p 0.98 \
    --max_planner_tokens 1024 \
    --max_questions -1 --n_threads 8 2>&1 | tee -a "$LOG"

echo "==== [A] extra-wide metrics ====" | tee -a "$LOG"
$PY scripts/compute_bestofn_metrics.py "$OUT_EW" K8_1stage_iter2_extrawidetemp 2>&1 | tee -a "$LOG"

##############################################
# STAGE B: apply selector v2-rows to extra-wide JSONL
##############################################
kill_vllm
echo "==== [B] launching selector v2-rows ====" | tee -a "$LOG"
$VLLM serve "$SEL_V2" \
    --served-model-name selector --port 8103 --dtype bfloat16 \
    --gpu-memory-utilization 0.85 --enforce-eager --max-model-len 8192 \
    > "${LOG}.serve_sel" 2>&1 &
wait_url http://localhost:8103/v1/models && echo "selector v2 READY" | tee -a "$LOG"

label="$(basename $OUT_EW .jsonl)_selectorV2rows"
echo "==== [B] $label ====" | tee -a "$LOG"
$PY scripts/compute_bestofn_with_selector.py \
    "$OUT_EW" "$label" --selector_host http://localhost:8103 --row_preview 2>&1 | tee -a "$LOG"

echo "==== ALL_DONE ====" | tee -a "$LOG"