© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-3-12B
    3. 31-GEMMASCOPE-2-RES-16K
    4. 8803
    Prev
    Next
    INDEX
    Explanations

    seeking confirmation

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

    The neuron lights up on discourse markers of personal stance or evaluation—words and small phrases that signal an opinion, judgment, or “meta” commentary (e.g. “deal,” “sounds,” “right,” “isn’t it,” “matters”).

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

    rhetorical questions or tag questions seeking agreement (like "right?", "isn't it?", or similar phrasings).

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

    colloquial, idiomatic expressions of subjective evaluation, often framed as tag questions or brief judgements.

    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_31_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
    போது
    0.55
     lectura
    0.54
    อาจ
    0.54
     islamic
    0.54
     पार्किंग
    0.53
    actually
    0.52
    人を
    0.50
     cittam
    0.50
    につき
    0.50
     vigueur
    0.49
    POSITIVE LOGITS
     right
    1.93
    吧
    1.90
    ですよね
    1.85
    了吧
    1.82
     그렇죠
    1.73
     Right
    1.66
    Right
    1.63
    right
    1.62
     그죠
    1.60
     RIGHT
    1.59
    Activations Density 0.196%

    No Known Activations