© 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. 47596
    Prev
    Next
    INDEX
    Explanations

    adverse, against, counter

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

    The neuron strongly activates on standalone heading or title words (i.e. the first token of section titles or paper/article titles).

    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
    enoord
    -0.79
    ඵ
    -0.77
     viable
    -0.76
    ühle
    -0.72
     Works
    -0.71
    नीय
    -0.70
    devtool
    -0.69
    継
    -0.69
    tings
    -0.69
     intermittently
    -0.69
    POSITIVE LOGITS
     contradicts
    1.03
    Противо
    0.90
     видел
    0.89
     opposes
    0.88
     resisted
    0.87
    stands
    0.87
     contradicted
    0.85
     Vrbo
    0.85
     rodeado
    0.85
     composée
    0.82
    Activations Density 0.008%

    No Known Activations