© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 48080
    Prev
    Next
    INDEX
    Explanations

    believe, trust, listen

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

    The neuron flags negation—especially the word “not” and its immediate context indicating a negative statement.

    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
     క
    -0.75
     ਦਾ
    -0.71
    logr
    -0.70
    zburg
    -0.68
    degrad
    -0.68
    던
    -0.68
    雪花
    -0.67
    yeon
    -0.66
     pojem
    -0.66
     thiệu
    -0.66
    POSITIVE LOGITS
     listen
    5.34
     listened
    4.81
     listening
    4.63
     listens
    4.53
    listen
    4.44
     Listen
    4.38
    Listen
    4.31
     Listening
    3.59
    listening
    3.56
    Listening
    3.41
    Activations Density 0.044%

    No Known Activations