evo2 SAE inference engine + CLI (encode / batch / serve)#1603
Draft
polinabinder1 wants to merge 1 commit into
Draft
evo2 SAE inference engine + CLI (encode / batch / serve)#1603polinabinder1 wants to merge 1 commit into
polinabinder1 wants to merge 1 commit into
Conversation
…e CLI + launch script PR1 of the inference/viz split. Importable Evo2SAE engine (no web deps): model+SAE load, encode (one sequence) + encode_batch (many, batched), generate/steer, feature_tracks. Thin FastAPI server + CLI (serve|batch|encode) + launch_inference.sh on top. The viz backend will import this later (PR2). Verified on a free GPU: single encode [S,15360], batched mixed-length encode, generate (plain + clamp + baseline). Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Contributor
|
Important Review skippedAuto reviews are disabled on this repository. Please check the settings in the CodeRabbit UI or the ⚙️ Run configurationConfiguration used: Path: .coderabbit.yaml Review profile: CHILL Plan: Pro Plus Run ID: You can disable this status message by setting the Use the checkbox below for a quick retry:
✨ Finishing Touches🧪 Generate unit tests (beta)
Comment |
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Standalone Evo2 + SAE inference engine + CLI (
evo2_sae_infer). Loads a base Evo2(truncated to a layer) and a trained SAE once, then turns DNA into SAE features. It has
no web/dashboard dependency — the feature-explorer dashboard is just one consumer of its
servemode.Usage
evo2_sae_infer encode --sequence ACGT…→ annotate one sequence → top features (JSON on stdout)evo2_sae_infer batch --fasta in.fa --out out.parquet→ many sequences → per-sequence top featuresevo2_sae_infer serve→ FastAPI server on:8001(the dashboard backend)from evo2_sae_infer import Evo2SAE—encode/encode_batch/feature_tracks/generate(autoregressive generation with additive SAE-feature clamping)scripts/launch_inference.sh {serve|encode|batch}with env-var configFiles (746 lines, 5 files)
evo2_sae_infer/core.py— theEvo2SAEengineevo2_sae_infer/server.py— thin FastAPI layerevo2_sae_infer/cli.py— serve / encode / batchscripts/launch_inference.shevo2_sae_infer/__init__.py🤖 Generated with Claude Code