© 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-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 11391
    Prev
    Next
    INDEX
    Explanations

    a head shake means "yes," and a nod means "no.cultural norms and specific word meanings

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

    The neuron responds to uncommon or “rare” tokens—especially proper names, foreign-language words, and other low-frequency content 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_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
    า
    1.04
     Ambedkar
    0.89
     indole
    0.86
     Bact
    0.84
    釭
    0.82
    ியார்
    0.80
     Goshen
    0.79
     Nawab
    0.79
     chamar
    0.79
    י
    0.78
    POSITIVE LOGITS
    led
    0.74
    ﻋ
    0.74
    оз
    0.73
     stepped
    0.72
    ax
    0.71
    LEG
    0.70
     rushed
    0.69
     høj
    0.69
    agua
    0.68
    og
    0.67
    Activations Density 0.000%

    No Known Activations