© 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. 117797
    Prev
    Next
    INDEX
    Explanations

    was followed by positive descriptions

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

    The neuron fires strongly on words and phrases conveying enthusiastic positive sentiment (e.g. “was,” “very,” “extremely,” “wonderful”) in customer testimonials.

    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
    lihood
    -1.26
    imetry
    -1.13
    rawn
    -1.08
    ogical
    -1.07
    onday
    -1.07
    panes
    -1.07
    stration
    -1.05
    byes
    -1.04
    ícias
    -1.03
    spra
    -1.02
    POSITIVE LOGITS
     helpful
    1.70
     great
    1.42
     able
    1.38
     supportive
    1.36
     responsive
    1.31
     gracious
    1.17
     very
    1.15
     instrumental
    1.15
     accommodating
    1.11
     attentive
    1.08
    Activations Density 0.011%

    No Known Activations