© 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. 31-GEMMASCOPE-2-RES-16K
    7. 8944
    Prev
    Next
    INDEX
    Explanations

    towards shared

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

    The neuron responds to uncommon, high–information‐content tokens—especially proper names, specific dates or numbers, and other rare words.

    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_31_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
     Domingo
    0.75
    Би
    0.68
    ър
    0.68
    𝒊
    0.67
     Viet
    0.67
    كب
    0.66
     Lég
    0.66
    
    0.66
     оказа
    0.66
     Horizon
    0.66
    POSITIVE LOGITS
     dielectric
    0.96
     streamwise
    0.80
     medir
    0.76
     conformal
    0.73
     discontinuities
    0.71
     permittivity
    0.70
     correto
    0.69
     tive
    0.68
    ்கலை
    0.68
    本人
    0.68
    Activations Density 0.000%

    No Known Activations