© 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. 10-GEMMASCOPE-RES-131K
    7. 127711
    Prev
    Next
    INDEX
    Explanations

    side, beside, alongside

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

    The neuron fires on words expressing proximity or accompaniment—especially “side” (and variants like “beside,” “next to,” etc.) that mark one entity alongside another.

    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
     clever
    -1.47
    how
    -1.46
    как
    -1.39
    ichtet
    -1.38
     cleverly
    -1.36
     hänen
    -1.36
    Activités
    -1.35
     tenemos
    -1.33
     Being
    -1.33
     nifty
    -1.32
    POSITIVE LOGITS
    relatively
    1.76
    ぽん
    1.66
     în
    1.57
     focuses
    1.54
    再一次
    1.51
     the
    1.45
    
    1.41
    又是一
    1.41
    很快就
    1.39
    失去了
    1.36
    Activations Density 0.010%

    No Known Activations