© 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. 7168
    Prev
    Next
    INDEX
    Explanations

    story follows

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

    The neuron fires strongly on the word “film” (and closely related markers in movie‐description passages), i.e. it detects occurrences of “film” in text.

    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
     bereiken
    -0.89
    ਹਿ
    -0.85
     pudieron
    -0.83
    frases
    -0.82
    死ぬ
    -0.79
     mógł
    -0.79
     prédé
    -0.78
     auger
    -0.76
    ulica
    -0.75
    ιο
    -0.75
    POSITIVE LOGITS
     follows
    2.61
     follow
    2.53
     tells
    2.31
     centers
    2.11
    Follow
    2.08
    centers
    1.91
    follow
    1.88
     centres
    1.87
     Follow
    1.85
     tell
    1.80
    Activations Density 0.133%

    No Known Activations