© 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. 22-GEMMASCOPE-RES-131K
    7. 33177
    Prev
    Next
    INDEX
    Explanations

    reporting issues or findings

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

    The neuron activates on instances of the verb “report” (and its variants like “reports,” “reported”) often in contexts giving statistics or findings.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    Questions
    -1.12
    ↵↵
    -1.09
    h
    -1.02
    "
    -1.02
    V
    -1.00
     —
    -0.96
    rophes
    -0.96
     preguntas
    -0.96
     I
    -0.94
    lamations
    -0.93
    POSITIVE LOGITS
     that
    1.28
     reported
    1.26
     parachoque
    1.18
     reporting
    1.17
    鼷
    1.15
     būs
    1.14
    Fakta
    1.10
     darbo
    1.07
     strane
    1.07
     on
    1.05
    Activations Density 0.104%

    No Known Activations