© 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. 5746
    Prev
    Next
    INDEX
    Explanations

    different types

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

    The neuron fires strongly on occurrences of the word “different” (in any casing).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    the word "different" when it appears before nouns or adjectives to indicate variation or contrast.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    explicit mentions of difference or contrast, signaling that things vary or are not the same.

    oai_token-act-pair · gpt-5Triggered 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
    н
    1.48
    AL
    1.22
    ن
    1.10
    un
    1.09
    ν
    1.07
    ure
    1.05
     oncoming
    1.05
    a
    1.05
    EN
    1.04
    n
    1.02
    POSITIVE LOGITS
    なった
    1.34
     типов
    1.32
    iation
    1.30
     tipos
    1.28
    시켜
    1.28
     characterizes
    1.23
     тип
    1.18
    MEs
    1.15
    특별시
    1.15
    types
    1.13
    Activations Density 0.052%

    No Known Activations