© 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. 107740
    Prev
    Next
    INDEX
    Explanations

    equivalent

    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
    )&&(
    -0.28
    \Component
    -0.28
    otech
    -0.24
    æľªæĪIJ
    -0.24
    logs
    -0.24
     marvelous
    -0.24
    pared
    -0.23
    нÑıÑĤ
    -0.23
    械
    -0.23
    å¨ĺ
    -0.23
    POSITIVE LOGITS
    ei
    0.31
    HAL
    0.27
    iy
    0.25
     trajectory
    0.24
    å±
    0.24
    è¾Ł
    0.24
    @endsection
    0.24
    åĻ¨
    0.24
     balloons
    0.24
    icum
    0.23
    Activations Density 0.389%

    No Known Activations