© 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. 60911
    Prev
    Next
    INDEX
    Explanations

    movie titles and authors

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

    The neuron strongly activates on multi‐token spans that correspond to the titles of books, movies, or other creative works (including the digits of publication or release years).

    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
     Պ
    -0.80
    ঈ
    -0.79
    ✻
    -0.76
    𝟎
    -0.75
     Mohamad
    -0.75
    clusive
    -0.74
    DN
    -0.73
    orit
    -0.72
     interesses
    -0.72
    起身
    -0.70
    POSITIVE LOGITS
     Bly
    0.89
    Flan
    0.85
     Flanagan
    0.82
     Hau
    0.77
     Vä
    0.76
     ghosts
    0.74
    Hau
    0.71
    嘱
    0.70
    espan
    0.70
     JPEG
    0.69
    Activations Density 0.031%

    No Known Activations