© 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. 20-TRANSCODER-HP
    7. 63777
    Prev
    Next
    INDEX
    Explanations

    Technical documents

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Minus
    -0.28
    ::↵
    -0.27
    arian
    -0.27
    åĻİ
    -0.26
    _minus
    -0.26
    avar
    -0.26
    ignal
    -0.26
     minus
    -0.26
     Happiness
    -0.26
    ü
    -0.25
    POSITIVE LOGITS
    .seek
    0.32
    stu
    0.27
    è¯ķ
    0.26
    brick
    0.26
     od
    0.26
    createForm
    0.25
    æīįè¡Į
    0.25
    çģ¾åĮº
    0.24
    æłĩ
    0.24
    usc
    0.24
    Activations Density 10.443%

    No Known Activations