© 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-2-2B
    5. 22-GEMMASCOPE-TRANSCODER-16K
    7. 11089
    Prev
    Next
    INDEX
    Explanations

    words associated with violence, weapons and law enforcement

    oai_token-act-pair · gemini-2.0-flash

    gun-related topics

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     تانيه
    -0.61
    kuuta
    -0.58
    urina
    -0.57
     geval
    -0.55
    optera
    -0.55
    HKEY
    -0.55
    iglie
    -0.54
     joindre
    -0.54
     createSlice
    -0.54
    yaszt
    -0.53
    POSITIVE LOGITS
    house
    0.63
    WebControls
    0.53
    top
    0.52
    новниш
    0.50
     Baton
    0.47
     Normdatei
    0.46
    Scrolled
    0.46
    master
    0.45
    boat
    0.44
    coo
    0.44
    Activations Density 2.682%

    No Known Activations