© 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. 57416
    Prev
    Next
    INDEX
    Explanations

    school, learning, and insurance

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

    The neuron steadily ramps up activation with each successive token, effectively tracking a token’s position in the document (later tokens produce higher activations).

    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
    fucking
    -1.05
     particules
    -1.04
    worldly
    -0.99
     UTA
    -0.96
     ит
    -0.95
     що
    -0.94
     datter
    -0.92
    wesenheit
    -0.90
     prosi
    -0.88
     FUCKING
    -0.88
    POSITIVE LOGITS
     students
    1.52
     district
    1.52
     student
    1.30
     districts
    1.29
     teacher
    1.23
    district
    1.22
     school
    1.20
     öğren
    1.18
     adults
    1.09
     learners
    1.09
    Activations Density 0.071%

    No Known Activations