© 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. 2445
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on the token “Chat,” marking instances where the text invites or offers a chat/online help interaction.

    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
     electrón
    0.59
     vídeo
    0.58
    microphone
    0.57
    Playback
    0.56
     getVideo
    0.55
    歌
    0.55
    视频
    0.54
    首歌
    0.54
    audio
    0.54
     ଅ
    0.54
    POSITIVE LOGITS
     Chat
    0.74
    Chat
    0.69
     chat
    0.64
    chat
    0.62
     prec
    0.59
    onu
    0.58
     채팅
    0.58
    idik
    0.57
    chats
    0.56
    bans
    0.55
    Activations Density 0.000%

    No Known Activations