© 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. 10-GEMMASCOPE-RES-131K
    7. 15458
    Prev
    Next
    INDEX
    Explanations

    numbers followed by media channels

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

    The neuron activates on standalone numeric tokens (and the words “Channel,” “One,” “Two,” etc.) used to denote TV or radio channel/station identifiers.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    P
    -1.73
    F
    -1.70
    S
    -1.65
    O
    -1.63
    A
    -1.54
    整個
    -1.51
     流れ
    -1.45
    馓
    -1.44
    either
    -1.44
    Kudos
    -1.43
    POSITIVE LOGITS
    ."
    1.80
    »,
    1.60
    ть
    1.59
    )$,
    1.59
    }$,
    1.56
    ],
    1.55
    ၟ
    1.51
    焄
    1.50
    }=
    1.48
    ₤
    1.47
    Activations Density 0.009%

    No Known Activations