© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 7320
    Prev
    Next
    INDEX
    Explanations

    options for, direct support, contrasting with, generation models, presenting it, dataset names

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

    This neuron fires strongly on the very first few words of a new paragraph or section break, marking the start of a new segment.

    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
     florist
    0.70
     palju
    0.69
     Exec
    0.68
    𝗲
    0.68
    vať
    0.66
     आश्चर्य
    0.66
     dużo
    0.65
     कपकेक
    0.64
     stunning
    0.62
     tired
    0.62
    POSITIVE LOGITS
    0.75
    0.74
                                
    0.71
    ৯
    0.71
    ।
    0.68
    POSTFIELDS
    0.68
            
    0.68
     NOMBRE
    0.67
    ISON
    0.65
                    
    0.64
    Activations Density 0.323%

    No Known Activations