© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 6802
    Prev
    Next
    INDEX
    Explanations

    notes at the end

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

    The neuron is detecting blank‐line or paragraph‐break tokens (i.e. newlines and section separators).

    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
    recieve
    1.20
    썻
    1.16
     শুধুমাত্র
    1.12
     देखील
    1.12
     এছাড়াও
    1.11
    utilizzo
    1.11
     পরবর্তীতে
    1.09
    の為
    1.09
    됬
    1.05
     advancements
    1.03
    POSITIVE LOGITS
     variously
    1.22
     doubtless
    1.21
     IMHO
    1.19
     (
    1.08
     (=
    1.08
     --
    1.07
     ;-)
    1.07
     insofar
    1.05
     hitherto
    1.02
     inasmuch
    1.02
    Activations Density 1.278%

    No Known Activations