© 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. 35-TRANSCODER-HP
    7. 39965
    Prev
    Next
    INDEX
    Explanations

    code snippets/urls

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_35.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.35.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    w
    -0.32
    ZA
    -0.30
    zman
    -0.29
    seau
    -0.28
    meric
    -0.28
    tion
    -0.28
    IRR
    -0.28
    z
    -0.28
    zure
    -0.28
    zte
    -0.28
    POSITIVE LOGITS
    OO
    0.32
    s
    0.28
    RK
    0.28
    Rx
    0.27
    r
    0.27
    N
    0.26
    T
    0.26
     rains
    0.26
    L
    0.25
    TMP
    0.25
    Activations Density 67.479%

    No Known Activations