© 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-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 9544
    Prev
    Next
    INDEX
    Explanations

    common links, transmission, travel time, digit place, R-squared, Sun, resolve conflicts, racism

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

    This neuron responds to infrequently seen or “rare” tokens—especially numeric values and unusual proper‐name tokens—highlighting out‐of‐vocabulary or low‐frequency items.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     téléphonique
    0.71
    <0xB1>
    0.67
    ει
    0.66
     déjeuner
    0.64
     प्रस्तुत
    0.64
     preuves
    0.63
    <0x85>
    0.63
     |”
    0.63
    <0x9D>
    0.61
    AVAILABLE
    0.61
    POSITIVE LOGITS
     expresión
    0.88
    čius
    0.87
     Tencent
    0.87
    downcase
    0.86
    rcode
    0.86
    ਕਾਰ
    0.84
     именем
    0.83
     Astros
    0.80
     substring
    0.80
     flamingo
    0.80
    Activations Density 0.000%

    No Known Activations