© 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. 8992
    Prev
    Next
    INDEX
    Explanations

    come

    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
    ming
    -0.28
    è¿ĺèĥ½
    -0.27
     væ
    -0.26
    éķ¿æķĪæľºåĪ¶
    -0.25
    èĤ¢ä½ĵ
    -0.25
     *,↵
    -0.25
    mr
    -0.24
    ungen
    -0.24
    hr
    -0.24
    axis
    -0.24
    POSITIVE LOGITS
    åľ¨è¿ĩ渡
    0.27
    azard
    0.27
    usi
    0.26
     creeping
    0.26
    \e
    0.25
     incompet
    0.25
    quine
    0.25
    grown
    0.24
     setC
    0.24
     seeding
    0.24
    Activations Density 0.891%

    No Known Activations