© 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. 31-GEMMASCOPE-2-RES-16K
    7. 2155
    Prev
    Next
    INDEX
    Explanations

    The neuron selectively activates on technical tokens and codes—acronyms, model numbers, error labels, or other alphanumeric identifiers common in software‐oriented 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_31_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
    ﻙ
    0.81
    ﺢ
    0.77
     Με
    0.75
     मोदी
    0.73
    ﻥ
    0.73
     Modi
    0.73
    ﻡ
    0.72
    ﺵ
    0.71
     crée
    0.71
     καὶ
    0.70
    POSITIVE LOGITS
    flashlight
    0.84
    classlist
    0.76
    chatroom
    0.75
    noopener
    0.72
    Jonathan
    0.71
    timeslots
    0.71
    studentid
    0.70
    getElementsBy
    0.69
    sockfd
    0.69
    queryset
    0.68
    Activations Density 1.149%

    No Known Activations