© 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. 25-TRANSCODER-HP
    7. 107636
    Prev
    Next
    INDEX
    Explanations

    news/technical snippets

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     CVE
    -0.32
    忧
    -0.28
    upertino
    -0.27
    .picture
    -0.25
    icamente
    -0.25
     precedence
    -0.24
     pena
    -0.24
    åIJĦä½į
    -0.24
    罪
    -0.24
     Polic
    -0.23
    POSITIVE LOGITS
    Associated
    0.26
     secretive
    0.26
     WHILE
    0.25
    strain
    0.25
    ák
    0.25
     menstrual
    0.24
     Luigi
    0.24
     Associated
    0.24
    Association
    0.24
    lient
    0.24
    Activations Density 0.993%

    No Known Activations