© 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. 4959
    Prev
    Next
    INDEX
    Explanations

    The neuron is highly tuned to detect the word “type,” especially in “type of” constructions.

    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
    æt
    0.91
    ationen
    0.86
    郸
    0.81
    ৈত্র
    0.80
     estaría
    0.80
    cg
    0.79
    س
    0.79
    Š
    0.79
    Así
    0.78
     relembrar
    0.78
    POSITIVE LOGITS
    ﻳ
    0.75
     Readable
    0.73
    ну
    0.72
     USER
    0.72
     Transparent
    0.69
    ﻧ
    0.68
     அரு
    0.67
    墜
    0.67
    謐
    0.66
     Автор
    0.66
    Activations Density 0.000%

    No Known Activations