© 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. 10-GEMMASCOPE-RES-131K
    7. 66360
    Prev
    Next
    INDEX
    Explanations

    pointing towards something

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

    The neuron fires on occurrences of the word “point” (and its inflected forms like “points” and “pointing”).

    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
     but
    -1.78
     jenem
    -1.73
     demselben
    -1.69
     jenen
    -1.68
     gutem
    -1.65
     hohem
    -1.62
    ƴ
    -1.59
     derselben
    -1.52
    nict
    -1.51
    薩
    -1.48
    POSITIVE LOGITS
    婓
    1.93
     pointing
    1.74
     kimdir
    1.70
    鮰
    1.65
    什么呢
    1.61
     rocha
    1.61
    㑹
    1.61
     esmeralda
    1.59
    毉
    1.57
    先生の
    1.55
    Activations Density 0.020%

    No Known Activations