© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 47400
    Prev
    Next
    INDEX
    Explanations

    follow along, print and play, watch list, earn while you learn, make or break

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

    The neuron fires on single, prominent words that serve as section or article headings—typically the first word of a title or heading.

    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
    つまり
    -1.17
    ötzlich
    -0.97
    がある
    -0.94
    や
    -0.92
    も
    -0.92
    甦
    -0.90
    ulism
    -0.90
    änden
    -0.89
    culum
    -0.89
    は
    -0.88
    POSITIVE LOGITS
    ng
    1.12
     your
    1.05
    ings
    1.02
    inge
    0.96
    久し
    0.94
    ("")]
    0.94
    able
    0.94
     fest
    0.91
    mg
    0.91
     Yourself
    0.91
    Activations Density 0.086%

    No Known Activations