© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 11820
    Prev
    Next
    INDEX
    Explanations

    merchantability and fitness

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

    The neuron responds strongly to tokens composed of uppercase letters or with an initial capital letter (e.g. “Amelia,” “AlterField,” “MERCHANTABILITY”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    $
    -3.47
    k
    -3.02
    Q
    -2.98
    =
    -2.91
    a
    -2.89
    ).
    -2.81
    -
    -2.75
    ent
    -2.72
    ]
    -2.64
    ing
    -2.63
    POSITIVE LOGITS
    瑨
    2.63
    乆
    2.61
    Ꮸ
    2.58
    ांकि
    2.56
    </u>
    2.55
    amkeit
    2.53
     algun
    2.52
    🦤
    2.47
    Ꮵ
    2.45
    ڧ
    2.45
    Activations Density 0.016%

    No Known Activations