© 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. 19-TRANSCODER-HP
    7. 7936
    Prev
    Next
    INDEX
    Explanations

    night sky

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_19.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.19.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ä¸īåĽĽ
    -0.25
    ä¸ĥåħ«
    -0.24
    åIJĮæŶ
    -0.24
     switched
    -0.23
     bound
    -0.23
     Sne
    -0.23
     seguint
    -0.23
     sweeping
    -0.22
     Spell
    -0.22
     starters
    -0.22
    POSITIVE LOGITS
    elix
    0.32
    igrated
    0.29
    /Library
    0.27
    åĩºçİ°éĹ®é¢ĺ
    0.27
    stderr
    0.25
     Kore
    0.25
    roman
    0.25
    管çIJĨä¸Ńå¿ĥ
    0.25
    å·¥å§Ķ
    0.25
    磶
    0.25
    Activations Density 0.015%

    No Known Activations