© 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. 48691
    Prev
    Next
    INDEX
    Explanations

    access and retrieve

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

    This neuron fires on first-person problem-statements—that is, tokens where the user refers to themselves (“I”, “I’m having trouble”, “I can…”, “I expect…”) and related auxiliary verbs describing their actions or difficulties.

    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
    glyphicon
    -0.85
    を入力
    -0.79
     calculator
    -0.78
    ᡴ
    -0.77
     Generate
    -0.77
    默认为
    -0.75
    тельность
    -0.75
    showLoading
    -0.75
    hunters
    -0.74
    getPhone
    -0.74
    POSITIVE LOGITS
     access
    2.66
     accessing
    2.19
    access
    2.06
     accesses
    2.05
    Access
    1.90
     accessed
    1.79
     Access
    1.78
    Accessing
    1.73
     retrieve
    1.61
     Zugriff
    1.59
    Activations Density 0.707%

    No Known Activations