© 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. 34-GEMMASCOPE-RES-131K
    7. 63652
    Prev
    Next
    INDEX
    Explanations

    warmth, weather, summer, cozy

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

    The neuron strongly activates on occurrences of the substring “warm” (including forms like “Warm,” “warmth,” and “warm-up”).

    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
    昻
    -0.98
     información
    -0.91
     século
    -0.91
     profilo
    -0.91
    isure
    -0.91
    сова
    -0.90
    跄
    -0.90
    ENDPOINT
    -0.88
     acidente
    -0.85
     somewhat
    -0.83
    POSITIVE LOGITS
    hearted
    1.27
     Warm
    1.27
     warm
    1.21
    onger
    1.20
     warms
    1.18
    warm
    1.12
    Warm
    1.11
    itamente
    1.01
     warmest
    0.99
     hearted
    0.96
    Activations Density 0.010%

    No Known Activations