© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 79072
    Prev
    Next
    INDEX
    Explanations

    takes damage or criticism

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

    The neuron fires on instances of “take/took/taken” used idiomatically to describe receiving or suffering a negative impact (e.g. “taken a beating,” “took a hit,” “taken plenty of flack”).

    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
    MENTS
    -0.82
     decorated
    -0.80
    ッター
    -0.77
    riscoll
    -0.76
    ностях
    -0.75
    ZONE
    -0.75
    شة
    -0.74
    anken
    -0.73
    venu
    -0.73
    zte
    -0.73
    POSITIVE LOGITS
     hits
    1.96
     punishment
    1.58
     hit
    1.58
    hits
    1.51
     brunt
    1.46
    Hits
    1.44
     knocks
    1.40
     Hits
    1.38
     abuse
    1.33
    hit
    1.19
    Activations Density 0.014%

    No Known Activations