© 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. 18-TRANSCODER-HP
    7. 88835
    Prev
    Next
    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_18.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.18.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     st
    -0.29
    ego
    -0.28
     origin
    -0.27
     hol
    -0.27
    forward
    -0.26
     sting
    -0.26
     tol
    -0.26
    aling
    -0.25
    eh
    -0.25
     favor
    -0.25
    POSITIVE LOGITS
    ernaut
    0.28
    åħ¨çIJĥæľĢ大
    0.26
    rients
    0.25
    readonly
    0.25
    .Click
    0.25
     unread
    0.24
     amassed
    0.24
    yte
    0.24
    æł¹æľ¬
    0.24
    mob
    0.24
    Activations Density 0.963%

    No Known Activations