© 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. 22-GEMMASCOPE-RES-131K
    7. 34636
    Prev
    Next
    INDEX
    Explanations

    negative reputation or perception

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

    The neuron detects words that signal reputation or social judgment (e.g. reputation, stereotypes, preconceived notions, labeled).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    حه
    -0.99
    が高
    -0.97
    <td>
    -0.93
     था
    -0.90
    ere
    -0.90
    gn
    -0.90
    bra
    -0.89
     Cuth
    -0.89
    POST
    -0.88
     chaleur
    -0.88
    POSITIVE LOGITS
     reputation
    1.28
     stereotypes
    1.23
     stereotype
    1.09
    のイメージ
    1.05
     perception
    1.01
    sterne
    1.01
     perceptions
    0.99
     synonymous
    0.96
     alleen
    0.96
     réputation
    0.94
    Activations Density 0.057%

    No Known Activations