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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 92511
    Prev
    Next
    INDEX
    Explanations

    effects and impacts

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

    The neuron activates specifically on numeric tokens (quantities, measurements, counts).

    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
     utilisent
    -0.92
    cension
    -0.90
    тическая
    -0.85
    蟊
    -0.85
     塗装
    -0.84
     semelhantes
    -0.84
     ایش
    -0.83
    edded
    -0.81
     kursi
    -0.81
     valable
    -0.81
    POSITIVE LOGITS
     effect
    7.25
     effects
    7.22
     impact
    6.22
    effect
    5.34
    effects
    5.28
     Effects
    5.22
     impacts
    5.16
     Effect
    5.06
    Effect
    5.00
    Effects
    4.88
    Activations Density 0.435%

    No Known Activations