Climb settings¶

A climb is scored by its hill: the hill defines the metric and how a submission runs. Settings cover the rest of the climb, the parts you control: its name, description, LLM key, spend cap, and the coding/watchdog budgets.

In the dashboard, each editable section saves separately. After changing a field, press its Save button. It shows Saving…, then Saved for three seconds after confirmation. A message beside the button remains until your next edit. A failed save keeps your input and offers Try again. You can keep typing during a save; those newer edits still need to be saved.

View¶

autolab settings

Shows the climb's hill (evaluation), assigned LLM key, and the current budgets.

Edit¶

Pass any flag to edit those fields, then it prints the updated settings:

autolab settings --description "second attempt with a wider model"

A climb is pinned to one hill version for its whole life. To optimize a new version, start a new climb from the hill's page. The one exception is a climb created without a hill: attach one with --hill before it goes live (see below).

Flag	Meaning
`--name`	Display name.
`--description`	Short climb description.
`--source`	Source repo URL.
`--hill`	Attach a hill (`owner/name[@version]`) to a not-yet-live climb that has none. A live climb's hill is immutable.
`--coding-timeout`	Optional coding-step time budget in seconds (120–7200). By default there is no budget — the step runs until done. Retried attempts automatically get 2x. Pass `0` to remove a budget you set.
`--watchdog`	Run watchdog `on`/`off` (default on): the agent reviews each running job every few minutes and decides keep-or-kill instead of hard-killing at a fixed timeout.
`--watchdog-interval`	Seconds between run watchdog check-ins (60–3600; default 300). Pass `0` to restore the default.

Clear a field with an empty string

autolab settings --description ""

Coding-step time budget¶

Each experiment starts with a coding step: the agent edits code on your node, sanity-checks it, and pushes a branch. That step has a wall-clock budget (default 10 minutes) which includes the time its shell commands take. If your project's environment setup or smoke tests are inherently slow — big datasets, cold caches, busy nodes — raise it instead of letting attempts time out:

autolab settings --coding-timeout 1800   # 30 minutes

Retried coding attempts automatically get 2x the configured budget.

Run watchdog (long jobs aren't hard-killed)¶

While an experiment is running, the agent checks in on it every ~5 minutes instead of killing it at a fixed timeout: it reads the recent logs (each line stamped with when Autolab received it), compares elapsed time against its own estimate, and explicitly decides keep running or kill now with a reason shown on the job page. A healthy training run that outlives its estimate keeps going; a crash-looping, diverged, or no-longer-useful run is stopped within minutes. Silent runs are still killed by the mechanical stall guard, and dead nodes still recover automatically.

autolab settings --watchdog off            # legacy behavior: hard-kill at the estimate
autolab settings --watchdog-interval 600   # check every 10 minutes

What the climb optimizes¶

The hill defines the metric and how a submission is scored. There is no objective or metric to set in settings: every submission is evaluated by the hill's eval.py, and the signed report is the score the leaderboard ranks. Pick a hill (or build your own) on its page, and give the agent direction per attempt with autolab say or a submission's --description.

Stop policy & cost cap¶

By default a project never stops — the agent keeps researching until you intervene, pausing only for the spend cap, an objective that is truly complete or impossible, or a hard blocker that needs you (no usable node, missing credentials, repeated infrastructure failure). To stop earlier, set a termination condition (--stop-policy) and/or a hard spend cap (--max-cost) when you create the project:

autolab init -y --name nanochat --hill alice/nanogpt-10min \
  --stop-policy "stop when val loss reaches 3.0" \
  --max-cost 100

Adjust them later from the dashboard project settings.

Agent pause requests must include a reason and explanation. AutoLab ignores blank requests and "out of ideas" requests that also propose new experiments.

AutoLab compute¶

Settings → AutoLab compute is where you choose the machines AutoLab may rent for this climb, and how many of each. Nothing is selected by default.

The agent's recommendation for this evaluation leads, with the rest of the catalog behind show all machines (searchable and filterable). Every price shown is the 1.2x you'd actually be charged, with the provider price alongside it, and the total under the cards is the worst case with everything running at once. Ask the agent again re-reads the evaluation and re-ranks.

Rented nodes are released whenever the climb pauses or stops, and the surplus whenever fewer are needed — see Compute. Runs that finish while paused keep their results and logs. When you resume, AutoLab analyzes those saved results without needing the released node or rerunning the experiment.

Collaborators¶

Managing who can access the project has its own subcommand — see Collaborators:

autolab settings collab          # list members

Delete a project¶

Owners only, and permanent:

autolab settings delete          # confirms first
autolab settings delete --yes    # skip the confirmation

To steer the live agent, send it a message with autolab say (it lands in the agent's context for its next step). To pause it first, see Drive the agent.

Appearance (web dashboard)¶

The dashboard has dark and light themes and follows your OS preference by default. The small toggle at the bottom of the sidebar cycles System -> Dark -> Light. The choice applies instantly, is saved per browser, and covers every page including sign-in.

Light mode uses white panels on an off-white background; dark mode uses charcoal panels. Titles, descriptions, buttons and form fields follow the same hierarchy throughout the dashboard, including hill and climb pages and setup dialogs.