© 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. 34530
    Prev
    Next
    INDEX
    Explanations

    taking part, shape, way

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

    The neuron fires on second-person instructional language—especially the pronoun “you” and nearby guiding words like “how,” “can,” and “this” in tutorial or how-to style sentences.

    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
     this
    -1.45
    <bos>
    -1.40
     these
    -1.31
     people
    -1.25
     went
    -1.16
     thinks
    -1.16
     begy
    -1.12
     porcent
    -1.12
     perché
    -1.12
     can
    -1.10
    POSITIVE LOGITS
     thereby
    1.37
    走吧
    1.34
     jednocześnie
    1.33
     momentous
    1.31
     każdej
    1.28
     estadia
    1.24
    だけではなく
    1.22
    perhaps
    1.20
     unveils
    1.20
     crucial
    1.19
    Activations Density 0.059%

    No Known Activations