© 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. 34-GEMMASCOPE-RES-131K
    7. 32802
    Prev
    Next
    INDEX
    Explanations

    image, growth, service, fire, height

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

    The neuron activates primarily on numeric literals (especially larger multi‐digit numbers) within code or configuration settings.

    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
     versátil
    -0.89
    AILABLE
    -0.84
     COMERC
    -0.82
     backless
    -0.82
    BIÉN
    -0.82
     collectionView
    -0.79
    Stochastic
    -0.79
    Avon
    -0.79
    ارع
    -0.77
    Tere
    -0.77
    POSITIVE LOGITS
    height
    0.84
    だったが
    0.79
    AutoSize
    0.75
     height
    0.75
     ISS
    0.74
    脚注の使い方
    0.74
    ڕ
    0.74
     heights
    0.73
     управление
    0.72
    proton
    0.71
    Activations Density 0.001%

    No Known Activations