© 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. 18-TRANSCODER-HP
    7. 89881
    Prev
    Next
    INDEX
    Explanations

    teacher

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_18.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.18.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     MTV
    -0.28
    EMENT
    -0.25
    ning
    -0.25
    ayah
    -0.25
    鬻
    -0.24
    lse
    -0.24
     spiritually
    -0.24
    .variant
    -0.24
    EEDED
    -0.24
    gettext
    -0.23
    POSITIVE LOGITS
    quake
    0.27
    .hy
    0.26
    ##↵↵
    0.26
    ãĤ¤
    0.25
     harmonic
    0.24
     hy
    0.24
     young
    0.24
    å·´
    0.23
     initials
    0.23
    _you
    0.23
    Activations Density 19.670%

    No Known Activations