© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 62677
    Prev
    Next
    INDEX
    Explanations

    movie releases

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

    The neuron fires on film‐title tokens (proper names like “Thor,” “Dark World,” “Black Panther,” etc.) and the surrounding release-info language (words such as “hitting,” “theaters,” “released,” dates, etc.).

    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
     water
    -0.97
     three
    -0.90
     two
    -0.90
     as
    -0.89
     four
    -0.88
     many
    -0.83
     problems
    -0.83
    lytes
    -0.81
     greatly
    -0.81
     used
    -0.81
    POSITIVE LOGITS
     release
    1.90
     theaters
    1.86
    Release
    1.81
    release
    1.76
     released
    1.63
     cinemas
    1.60
     Release
    1.53
    released
    1.52
     theatres
    1.51
     RELEASE
    1.46
    Activations Density 0.038%

    No Known Activations