© 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. 7828
    Prev
    Next
    INDEX
    Explanations

    converted

    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
    设æĸ½å»ºè®¾
    -0.29
    è§ļ
    -0.28
    ennis
    -0.27
     wg
    -0.25
    endency
    -0.24
    èĬ±åĽŃ
    -0.24
    åħį责声æĺİ
    -0.24
     MJ
    -0.24
    .inline
    -0.23
     nothing
    -0.23
    POSITIVE LOGITS
    ARNING
    0.28
     mon
    0.28
    /support
    0.27
     bin
    0.26
     Hel
    0.26
     quadr
    0.26
     Heat
    0.26
     apo
    0.26
     hel
    0.25
    æĶ¯æŁ±
    0.25
    Activations Density 0.100%

    No Known Activations