© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 1242
    Prev
    Next
    INDEX
    Explanations

    The neuron is essentially a “reading‐position” detector—it fires more strongly on tokens that appear progressively later in the document, encoding how far you are through the text.

    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
     ragazza
    0.43
     girlfriend
    0.42
     girl
    0.41
     プレゼント
    0.40
     satisfied
    0.38
     Chocol
    0.38
     Girlfriend
    0.38
     俺
    0.38
     девушка
    0.38
     Fridge
    0.37
    POSITIVE LOGITS
     counterv
    0.37
     tái
    0.36
     compensatory
    0.33
     অন্যান্য
    0.32
     شركات
    0.32
     regulatory
    0.31
     obstructions
    0.31
     മറ്റ്
    0.31
     antioxidants
    0.31
    ೮
    0.30
    Activations Density 0.000%

    No Known Activations