© 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. 13609
    Prev
    Next
    INDEX
    Explanations

    5 elements

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

    This neuron isn’t responding to any linguistic feature of the words—it simply fires on whatever token lands at a particular fixed position in its input window, regardless of its meaning.

    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
    後ろ
    0.73
    𝙧
    0.73
    を経
    0.69
    יות
    0.68
    ુ
    0.67
    dozen
    0.66
    say
    0.66
    ી
    0.66
    vertebr
    0.65
    شناسی
    0.65
    POSITIVE LOGITS
     Hungry
    0.86
     shitty
    0.82
     Tide
    0.81
     Sverige
    0.80
     компанией
    0.80
     Folge
    0.79
     Persön
    0.79
     Bedingungen
    0.79
     Timm
    0.79
     Dame
    0.78
    Activations Density 0.000%

    No Known Activations