Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 8631
    Prev
    Next
    INDEX
    Explanations

    feature selection

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     feature
    -1.09
     features
    -1.09
    FEATURES
    -0.88
    features
    -0.84
     Features
    -0.76
    ABB
    -0.76
    Feature
    -0.73
    rollable
    -0.73
    зможно
    -0.73
    íts
    -0.72
    POSITIVE LOGITS
     extraction
    1.30
     extractor
    1.09
     vector
    1.05
     engineering
    1.05
     space
    1.02
     selection
    1.00
     vectors
    0.96
    vector
    0.94
    Extraction
    0.93
     Space
    0.92
    Activations Density 0.022%

    No Known Activations