© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 14451
    Prev
    Next
    INDEX
    Explanations

    width:

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

    The neuron fires on numeric literals (especially CSS dimensions or other numbers) in the document.

    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
     periods
    -1.21
     when
    -1.16
     these
    -1.15
    ↵
    -1.15
    }
    -1.12
     に関する
    -1.04
    並ぶ
    -1.03
     being
    -1.02
    ますね
    -1.02
     having
    -1.00
    POSITIVE LOGITS
    ETY
    1.23
    ⒉
    1.21
    ᚴ
    1.19
    軀
    1.13
    gebühr
    1.10
    ₳
    1.09
    ваете
    1.08
    crollView
    1.08
    beitung
    1.07
    咣
    1.07
    Activations Density 0.005%

    No Known Activations