© 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. 33-TRANSCODER-HP
    7. 53327
    Prev
    Next
    INDEX
    Explanations

    Code and text strings

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    om
    -0.31
    çĽĸ
    -0.28
    oda
    -0.27
    leta
    -0.27
    èĵĭ
    -0.26
    cá
    -0.26
    RIX
    -0.26
    ocy
    -0.25
     TOR
    -0.25
    Connell
    -0.25
    POSITIVE LOGITS
    hips
    0.29
    ees
    0.28
    iry
    0.26
    eenth
    0.26
     Español
    0.25
    urf
    0.24
    ibur
    0.24
    .Cursors
    0.24
    太快
    0.24
     Hoy
    0.24
    Activations Density 0.008%

    No Known Activations