© 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. 16156
    Prev
    Next
    INDEX
    Explanations

    The neuron flags the very first (non‐punctuation) content word at the start of a new segment or sentence.

    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
     Cette
    0.84
    ﻮ
    0.83
    y
    0.82
    ский
    0.80
     Selon
    0.79
    ariya
    0.77
    gruppe
    0.77
     Menurut
    0.77
    gruppen
    0.77
     黃
    0.76
    POSITIVE LOGITS
     prostu
    0.84
    pathetic
    0.82
     cashless
    0.75
    बिज
    0.75
     idempotent
    0.74
    পণ
    0.73
    שוט
    0.73
     docile
    0.72
     tropos
    0.72
     cupid
    0.71
    Activations Density 0.000%

    No Known Activations