© 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. 58625
    Prev
    Next
    INDEX
    Explanations

    information and details

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

    The neuron flags first-person plural/editorial “we” statements and similar introductory or guiding phrases (e.g. “we all know,” “we are going to talk about,” “let’s have a look at,” etc.).

    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
     during
    -1.03
     in
    -0.96
     for
    -0.96
     bereits
    -0.89
     bluish
    -0.88
     after
    -0.88
     with
    -0.87
     without
    -0.87
     sehr
    -0.86
     greenish
    -0.86
    POSITIVE LOGITS
    information
    1.16
     médi
    1.14
    Internet
    1.14
    Information
    1.10
     information
    1.09
     publicly
    1.09
     اطلاعات
    1.09
     details
    1.08
    details
    1.07
    anything
    1.06
    Activations Density 0.007%

    No Known Activations