© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 34401
    Prev
    Next
    INDEX
    Explanations

    film titles with directors

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

    The main thing this neuron does is find numeric year tokens (years in parentheses) in filmography entries.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    を越
    -1.13
     recente
    -1.11
     who
    -1.10
    のではないでしょうか
    -1.05
     vogel
    -1.03
     noten
    -1.03
     jeudi
    -1.00
    })->
    -1.00
    のですか
    -0.99
     カット
    -0.99
    POSITIVE LOGITS
     When
    1.40
     There
    1.38
     with
    1.36
     but
    1.27
     there
    1.19
     entendido
    1.15
     around
    1.06
     Although
    1.05
     on
    1.05
    camisa
    1.05
    Activations Density 0.019%

    No Known Activations