© 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. Qwen3-4B
    5. 19-TRANSCODER-HP
    7. 8766
    Prev
    Next
    INDEX
    Explanations

    referencing something specific

    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
    ç»ıèIJ¥èĮĥåĽ´
    -0.25
    áºŃu
    -0.24
    reaction
    -0.24
    stakes
    -0.24
    rens
    -0.24
    ÑĪив
    -0.24
    ekyll
    -0.23
    çķ¬
    -0.23
    eeper
    -0.23
    串
    -0.23
    POSITIVE LOGITS
     manipulate
    0.26
    è¿Ŀ
    0.26
    åIJĽ
    0.25
     misconduct
    0.24
    æĥ
    0.24
    hard
    0.24
     manipulated
    0.23
    геÑĢ
    0.23
    noch
    0.23
    ableObject
    0.23
    Activations Density 0.019%

    No Known Activations