© 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. 11709
    Prev
    Next
    INDEX
    Explanations

    This neuron activates most strongly on tokens that occur at the ends of input segments, flagging end‐of‐chunk positions.

    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
    écies
    0.79
    ক্ট
    0.78
    ാൻഡ്
    0.77
    ۂ
    0.76
    क्रेट
    0.75
     बिज़
    0.72
     కుటుంబ
    0.70
    ంధ
    0.70
    ించిన
    0.70
    auso
    0.69
    POSITIVE LOGITS
    1.08
    Unary
    0.72
    にし
    0.70
                    
    0.68
    お
    0.68
    0.67
     Parallel
    0.66
    すべての
    0.66
    不
    0.65
     держав
    0.64
    Activations Density 0.000%

    No Known Activations