© 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
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 15336
    Prev
    Next
    INDEX
    Explanations

    interesting and

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

    This neuron is looking for instances of subjective evaluation cues—most notably the word “interesting.”

    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
    ל
    1.03
    л
    0.97
    াল
    0.95
    ल
    0.92
    Thankfully
    0.89
    к
    0.88
     Thankfully
    0.86
     Expectations
    0.85
    ра
    0.84
    ল
    0.82
    POSITIVE LOGITS
     parallels
    0.95
     🤔
    0.84
    🤔
    0.83
     vértice
    0.81
     analogies
    0.79
    Skull
    0.79
    ség
    0.78
    ün
    0.77
    एँ
    0.77
     contradictions
    0.77
    Activations Density 0.131%

    No Known Activations