tp sweep - TunePlane

tp sweep <EXP> -g KEY=V1,V2 [options]

Each variant is one standard submission (quotas / queuing apply as usual). See Hyperparameter Sweeps for the workflow walkthrough.

Option	Description
`-g, --set-grid KEY=V1,V2` (required, repeatable)	Grid axis; multiple axes take the Cartesian product
`-s, --set KEY=VALUE`	Fixed overrides shared by all variants, repeatable
`--profile`	Resource declaration, same as `tp submit`
`-m, --method`	Method id; if omitted, read from the lock file
`--dry-run`	Only print the variant list, do not submit
`--allow-dirty` / `--no-validate`	Same as `tp submit`

Examples

# 3 × 2 = 6 variants; review first, then submit
tp sweep my-grpo --profile h200:4 \
  -g policy.optimizer.kwargs.lr=1e-6,2e-6,5e-6 \
  -g grpo.kl_coef=0.01,0.05 \
  --dry-run
tp sweep my-grpo --profile h200:4 \
  -g policy.optimizer.kwargs.lr=1e-6,2e-6,5e-6 \
  -g grpo.kl_coef=0.01,0.05

# Stop the whole group with one command
tp job stop-sweep <SWEEP_ID>

`tp sweep`

Submit a hyperparameter sweep

tp sweep <EXP> [options]

Argument	Description
`EXP`	Experiment name or path (required)

Option	Description
`--set-grid`, `-g`	A grid axis, repeatable; several axes form a cartesian product. e.g. -g policy.lr=1e-5,2e-5 -g grpo.kl=0.01,0.05 (required · repeatable)
`--set`, `-s`	A fixed override shared by every variant, repeatable (repeatable)
`--profile`	Target hardware and resources in one argument: h200 (the registry default shape), h200:4 (4 GPUs), h200:16 (2 full nodes). Repeat it to split pools per role (the heterogeneous extension point): —profile train=h200:8 —profile rollout=h100:2. Comma-separate roles that share a pool: —profile actor,rollout=h200:8 —profile teacher=h200:8 (repeatable)
`--method`, `-m`	Method identifier; read from the experiment’s recipe.lock.json when omitted
`--allow-dirty`	Allow uncommitted changes in the working tree
`--no-validate`	Skip the pre-submission validation
`--dry-run`	Only print the variant list, submit nothing
`--early-stop`	Enable ASHA early stopping: cut the clearly hopeless variants at a few step rungs
`--stop-metric`	Which metric early stopping watches; defaults to the method’s declared primary metric (see tp methods <method>)
`--rungs`	Early-stopping rungs (steps), comma separated. Defaults to 100,300,900
`--keep` `<int range>`	Keep 1/keep of the variants at each rung. 3 is the usual value: cutting harder hurts slow starters (default `3`)