© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 91897
    Prev
    Next
    INDEX
    Explanations

    redheaded, eyed, faced, legged, voiced

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron detects color descriptors (adjectives naming colors) modifying nearby nouns.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
    atier
    -0.78
     infest
    -0.76
    bilis
    -0.75
    pecabezas
    -0.74
    nivel
    -0.74
     ratified
    -0.73
     imprimé
    -0.73
     тему
    -0.73
    暟
    -0.72
     Italij
    -0.72
    POSITIVE LOGITS
     haired
    2.23
     eyed
    2.17
     faced
    2.05
    eyed
    1.70
    faced
    1.58
     headed
    1.50
    haired
    1.48
     legged
    1.47
     voiced
    1.44
     minded
    1.30
    Activations Density 0.035%

    No Known Activations