© 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. 34-GEMMASCOPE-RES-131K
    7. 76934
    Prev
    Next
    INDEX
    Explanations

    we followed by research verbs

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

    This neuron responds to first-person/discourse markers and reporting verbs commonly used in scientific abstracts (e.g. “Our,” “we,” “show,” “report”).

    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
    してほしい
    -0.91
    Cassie
    -0.89
     researched
    -0.83
    してみました
    -0.79
    tellte
    -0.79
    络
    -0.79
     arvio
    -0.78
     anuncia
    -0.77
     bilim
    -0.75
    ressão
    -0.75
    POSITIVE LOGITS
     studies
    1.06
     character
    1.04
     elegant
    1.02
     functional
    0.91
     functionally
    0.90
     characterizing
    0.89
     trabajado
    0.89
    character
    0.88
    機能
    0.85
    定位
    0.85
    Activations Density 0.483%

    No Known Activations