Extending Weird Generalization and Inductive Backdoors: New Ways to Corrupt LLMs
View KGrewal1/LearningBackdoors on GitHubExtending Weird Generalization and Inductive Backdoors: New Ways to Corrupt LLMs
RepoWorth models the current development value represented by the visible repository evidence. It does not include ideation, intellectual property, business value, product discovery, or sale price.
Modeled range $282.12 to $423.18
3,678 source lines across 12 authored files
The headline RepoWorth and measured source facts are ready. Comparable history, accepted changes, and badge evidence will appear here as enrichment finishes.
Ranked from the public GitHub contributor evidence available at the latest scan. This table describes measured repository activity, not ownership or original authorship.
| Rank | Contributor | Contributions | Share of top group | Latest measured activity |
|---|---|---|---|---|
| #1 |  KGrewal1@KGrewal1 | 33 | 100.0% | Current GitHub contributor snapshot |
Extending Weird Generalization and Inductive Backdoors: New Ways to Corrupt LLMs
These values come from the same immutable repository scan as the headline RepoWorth. They stay tied to the scanned commit so the reading can be inspected against one exact source state.
Established based on deterministic repository structure and complexity signals.
First-party implementation tokens eligible for the modeled code-scale input.
0 detected test files contribute separately from production code.
1 manifests and 0 workspace boundaries were detected.
1 documentation files remain visible as a distinct enablement signal.
36 dependency declarations and 0 public-interface signals were measured.
0 large files and 337 branch signals add structural context.
0 generated files were identified. Ignored and generated material does not inflate RepoWorth.
Code scale, branching, dependencies, languages, structure, and visible test evidence shape the implementation and verification effort represented by this repository.
Measured from the current source archive.
Derived from branch density, dependencies, languages, and code scale.
Measures visible test code relative to the source corpus.
This view separates authored evidence from the implementation, integration, verification, and engineering oversight represented by the current repository.
Reading boundary: These are modeled development inputs for the visible code. They do not include product discovery, company value, private systems, or commercial outcomes.
CodeValuation.com counted supported authored source on commit 8ef3768dca49095eb587227e61e12ca7b56ad4ef, measured deterministic repository signals, then estimated implementation, rework, testing, integration, and engineering oversight.
Supported first-party source, tests, configuration, manifests, repository structure, dependency signals, and framework boundaries.
Common generated and vendored files, build output, dependencies, lockfiles, minified assets, binaries, media, and private systems outside this repository.
Ideation, intellectual-property rights, security posture, production reliability, company value, sale price, revenue, user value, proprietary data, or the cost of discovering the original product.
This is a modeled current development valuation for the repository's visible code at a specific commit. It reflects code volume, implementation output, complexity, and engineering oversight. It is not a business valuation, investment appraisal, security review, or estimate of sale price, ideation, intellectual property, revenue, users, brand, proprietary data, undocumented domain knowledge, or private infrastructure.
Enter a public GitHub URL to create a current repository reading.