Remove hardcoded goodfire entity from SPDAdapter#467
Open
Conversation
SPDAdapter hardcoded "goodfire/spd/runs/{run_id}" when constructing
wandb paths, making the harvest pipeline unusable for other wandb
entities. Now accepts the full wandb_path from the config instead.
Antovigo
commented
Apr 3, 2026
| method config from the harvest DB (which is always populated before downstream | ||
| steps like autointerp run). | ||
| """ | ||
| if decomposition_id.startswith("s-"): |
Collaborator
Author
There was a problem hiding this comment.
I'm not 100% sure what these lines were for. Are there special cases when one might want to use a SPDAdapter without running the harvest first?
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description
SPDAdapternow accepts a fullwandb_pathinstead of just a run ID, removing the hardcodedgoodfire/spd/runs/prefix.Motivation and Context
The harvest pipeline was unusable for anyone with a different wandb entity because
SPDAdapter.__init__only took a run ID and hardcodedgoodfire/spd/runs/{run_id}. Now the full wandb path flows through from the harvest config.How Has This Been Tested?
make checkpasses (basedpyright + ruff)Does this PR introduce a breaking change?
No.
adapter_from_configstill works the same way (readswandb_pathfromSPDHarvestConfig).adapter_from_idnow always goes through_load_method_config, which reads the full config from the harvest DB. This requires harvest data to exist, butadapter_from_idis only called from downstream steps (autointerp, intruder eval, graph_interp) that already depend on harvest.