© 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. 53324
    Prev
    Next
    INDEX
    Explanations

    technical and scientific contexts

    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
     surface
    -0.28
    geb
    -0.27
    su
    -0.27
    size
    -0.26
    表éĿ¢
    -0.26
     zam
    -0.26
     mixed
    -0.26
    Z
    -0.26
     anth
    -0.25
    _bn
    -0.25
    POSITIVE LOGITS
    heat
    0.30
    uters
    0.27
     Heat
    0.26
    utan
    0.26
    ãĥĥãĥĪ
    0.26
    åĶ¾
    0.26
    ivirus
    0.25
    cratch
    0.25
    taxonomy
    0.25
     homo
    0.25
    Activations Density 0.015%

    No Known Activations