© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 76276
    Prev
    Next
    INDEX
    Explanations

    Arkansas, licenses, reserved copyrights

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

    The neuron fires on numeric tokens and literals—including version numbers, numeric constants, and digits in code.

    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
     electronically
    -0.83
    navn
    -0.78
    omotor
    -0.76
     plaine
    -0.75
    QWidget
    -0.74
     fres
    -0.74
    康
    -0.71
    othea
    -0.71
    "]:
    -0.70
    xem
    -0.70
    POSITIVE LOGITS
    Gdx
    1.83
     Arkansas
    1.70
    badlogic
    1.66
    Arkansas
    1.47
    gdx
    1.47
    cpy
    1.21
     Fayetteville
    1.20
     Shreveport
    1.13
     Ozark
    1.10
     Huntsville
    1.05
    Activations Density 0.012%

    No Known Activations