© 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. 10-GEMMASCOPE-RES-131K
    7. 111684
    Prev
    Next
    INDEX
    Explanations

    modal verbs followed by actions

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

    The neuron fires on second-person directive or invitational language—especially on modal verbs (can, could, might, want) and related cues when addressing “you.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     himself
    -1.71
     herself
    -1.45
     vecchia
    -1.34
     coloro
    -1.28
     two
    -1.27
     philanthropist
    -1.27
    himself
    -1.24
    --"
    -1.20
     două
    -1.20
    這些
    -1.19
    POSITIVE LOGITS
     themselves
    1.53
    ,
    1.36
     I
    1.33
     leurs
    1.31
     conjunta
    1.29
    阋
    1.23
     their
    1.22
    各自
    1.20
    HINT
    1.20
     Cost
    1.19
    Activations Density 0.021%

    No Known Activations