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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 23891
    Prev
    Next
    INDEX
    Explanations

    vehicle color modification

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

    The neuron is specialized in picking up language around selecting or changing colors (e.g. “supports,” “multiple colors,” “select,” “changing the color,” etc.).

    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
    ᴼ
    -0.88
    vádz
    -0.84
    íti
    -0.79
    Mär
    -0.79
     gradova
    -0.79
    Algorithms
    -0.78
    ()",
    -0.77
    اسة
    -0.77
    āda
    -0.76
    inus
    -0.75
    POSITIVE LOGITS
     modifications
    1.09
     paint
    1.09
     vehicle
    0.98
     reiterated
    0.86
     modification
    0.86
     modif
    0.84
     paints
    0.82
     tiefer
    0.81
     alterations
    0.80
     vehicles
    0.79
    Activations Density 0.002%

    No Known Activations