© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 19-TRANSCODER-HP
    7. 5302
    Prev
    Next
    INDEX
    Explanations

    prepositions

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_19.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.19.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    _MPI
    -0.27
    è·ª
    -0.27
    ç¦ģå¿Į
    -0.26
    (ci
    -0.24
     INA
    -0.24
    emark
    -0.24
     obstruct
    -0.23
    ени
    -0.23
    èĢĨ
    -0.23
    blick
    -0.23
    POSITIVE LOGITS
    own
    0.28
     GiriÅŁ
    0.28
     turno
    0.25
    ä½łèĩªå·±
    0.25
    icine
    0.24
    ä¾Ŀçħ§
    0.24
    ç¾İåĽ½æĢ»ç»Ł
    0.24
    ainer
    0.23
     ride
    0.23
    å¾Ĺå¤ļ
    0.23
    Activations Density 0.065%

    No Known Activations