© 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. 32-TRANSCODER-HP
    7. 43384
    Prev
    Next
    INDEX
    Explanations

    Tu

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    åıijæĶ¾
    -0.30
     thou
    -0.29
    sWith
    -0.26
    pets
    -0.26
     gratis
    -0.25
    over
    -0.25
    fra
    -0.25
    泪水
    -0.25
    /items
    -0.25
    _MARKER
    -0.25
    POSITIVE LOGITS
     strm
    0.27
     Crown
    0.26
    夷
    0.26
     debating
    0.26
    iling
    0.26
    ä¸į该
    0.25
     pumping
    0.25
    ying
    0.25
    ivi
    0.24
     Ways
    0.24
    Activations Density 0.000%

    No Known Activations