© 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. 65425
    Prev
    Next
    INDEX
    Explanations

    about people and relationships

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

    The neuron is triggered by multiword descriptive clauses—especially participial (“-ing”) phrases and prepositional attachments—picking up the verb, its prepositions, and their noun or pronoun complements.

    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
     môžete
    -1.06
     widać
    -1.05
     rarement
    -1.04
     pomocou
    -1.04
    骛
    -1.02
    usually
    -1.01
     praticamente
    -1.00
     reportedly
    -1.00
     žen
    -0.98
    ೨
    -0.97
    POSITIVE LOGITS
     instead
    1.27
     someone
    1.24
     just
    1.23
     or
    1.20
     only
    1.17
     somehow
    1.08
     although
    1.07
     another
    1.03
     some
    1.02
     could
    1.01
    Activations Density 0.027%

    No Known Activations