© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 38645
    Prev
    Next
    INDEX
    Explanations

    book titles and subject descriptions

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

    The neuron is keyed to the technical, feature-rich language of e-commerce product listings (e.g. “mattress,” “coil spring,” “layers,” “support,” “comfort”), flagging sections that read like marketing specifications.

    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
     receptacles
    -0.80
     wrongdoing
    -0.76
    arse
    -0.76
    всегда
    -0.75
     vídeos
    -0.73
    Dons
    -0.73
     dotyczące
    -0.71
     použí
    -0.71
    Voiced
    -0.71
     hanem
    -0.69
    POSITIVE LOGITS
     Rauch
    0.88
     البريط
    0.85
    pozycji
    0.82
    dding
    0.81
    метри
    0.80
    Pfalz
    0.79
    hütte
    0.79
    äste
    0.79
    ीक
    0.78
     selfie
    0.78
    Activations Density 0.015%

    No Known Activations