Buckets:
| # Queue 4 (no gates anywhere): MQ (hard answer diffusion) -> P2s (soft-token slots, fixed weight 1) -> MQ-soft -> E5-proc. | |
| cd /workspace/think-decide | |
| export PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True | |
| python src/mq.py train --out runs/mq > logs/mq_train.log 2>&1 | |
| python src/mq.py eval --model runs/mq/model.safetensors --out runs/mq/test --compile > logs/mq_eval.log 2>&1 | |
| scripts/run_proc_arm.sh p2s --n-think 32 --k-choices 1,2,4,8 --budget 36000 --fb-mode soft --gate-init 1.0 --gate-fixed > logs/p2s_run.out 2>&1 | |
| python src/mq.py train --out runs/mqsoft --soft > logs/mqsoft_train.log 2>&1 | |
| python src/mq.py eval --model runs/mqsoft/model.safetensors --out runs/mqsoft/test --soft-k 4 --compile > logs/mqsoft_eval.log 2>&1 | |
| scripts/run_e5_proc.sh > logs/p5_run.out 2>&1 | |
| echo QUEUE4_DONE > logs/queue4.done | |
Xet Storage Details
- Size:
- 824 Bytes
- Xet hash:
- c49be04bbaa941827a874fa92244ec3e95860263a03de0785e52c64997b46593
·
Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.