© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 38634
    Prev
    Next
    INDEX
    Explanations

    reported speech

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

    The neuron specializes in spotting reporting or attribution verbs (e.g. said, says, added) that introduce quoted speech.

    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
     at
    -1.17
     occasionally
    -1.09
     for
    -1.09
     all
    -1.08
     any
    -1.05
     to
    -1.05
     become
    -1.01
     maybe
    -1.01
     possibly
    -1.00
     no
    -0.97
    POSITIVE LOGITS
    ago
    1.18
     renfer
    1.13
     possibilités
    1.11
     WENT
    1.06
    ifice
    1.03
     ous
    1.02
    introduction
    1.01
     cial
    1.00
     theses
    1.00
     rezept
    1.00
    Activations Density 0.007%

    No Known Activations