© 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. 10859
    Prev
    Next
    INDEX
    Explanations

    cause/since clauses

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

    The neuron fires on discourse-linking cue words (especially sentence-initial connectors like “Because,” “Therefore,” “This,” etc.) that mark causal or logical transitions.

    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
    它
    1.32
    因此
    1.16
    まぁ
    1.02
    𝘤
    1.02
    𝘭
    0.98
    由于
    0.97
     suficiente
    0.97
    𝘵
    0.97
    𝘬
    0.97
    𝘳
    0.96
    POSITIVE LOGITS
     NTIS
    0.97
    se
    0.95
     nourishment
    0.95
    ah
    0.93
    er
    0.92
    ა
    0.91
    k
    0.89
    ce
    0.87
    u
    0.87
     nurturing
    0.87
    Activations Density 8.027%

    No Known Activations