© 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. 63865
    Prev
    Next
    INDEX
    Explanations

    moving home soon

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

    The neuron strongly activates on direct reader address—especially second-person pronouns (“you,” “your”) and related call-to-action phrasing.

    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
     أيضا
    -1.47
     остается
    -1.11
     кроме
    -1.11
    ülen
    -1.11
     zippers
    -1.10
     grayish
    -1.05
    ⃣
    -1.04
    ൺ
    -1.03
    小さく
    -1.02
    recognize
    -1.02
    POSITIVE LOGITS
    Whilst
    1.64
     classed
    1.56
     your
    1.55
     whilst
    1.49
    Firstly
    1.45
    Alongside
    1.45
     Whilst
    1.45
     swat
    1.24
     envisage
    1.23
     Firstly
    1.20
    Activations Density 0.049%

    No Known Activations