© 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. 42899
    Prev
    Next
    INDEX
    Explanations

    code snippets, documentation

    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
    åIJ¯è¿ª
    -0.28
    blings
    -0.27
    éĥĬ
    -0.27
    åıijå±ķçIJĨ念
    -0.27
     Colum
    -0.26
    isors
    -0.26
    @admin
    -0.26
    urement
    -0.25
    wares
    -0.25
     colum
    -0.25
    POSITIVE LOGITS
     phủ
    0.26
    èĥª
    0.24
    .GetObject
    0.24
    emon
    0.24
    inking
    0.23
    elve
    0.23
     post
    0.23
     phá»ķ
    0.23
    inen
    0.23
    æĭĪ
    0.23
    Activations Density 0.001%

    No Known Activations