© 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. 37959
    Prev
    Next
    INDEX
    Explanations

    talking points

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

    The neuron fires when it sees the standalone word “Talking,” typically as a title or label.

    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
     Pore
    -0.69
    arkov
    -0.67
    tamen
    -0.66
     loosing
    -0.64
    配套
    -0.64
    filled
    -0.63
     Allgeme
    -0.63
    benhavn
    -0.63
    ative
    -0.63
     mounts
    -0.63
    POSITIVE LOGITS
    calientes
    0.69
    ASC
    0.69
    wein
    0.68
    岬
    0.67
     Vad
    0.66
    GUILayout
    0.65
    Vad
    0.64
    úc
    0.64
    了一口气
    0.64
     croire
    0.63
    Activations Density 0.036%

    No Known Activations