© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 19-TRANSCODER-HP
    7. 6581
    Prev
    Next
    INDEX
    Explanations

    URLs and code

    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
     scanning
    -0.28
     succeeded
    -0.28
    åĢĴæĺ¯
    -0.25
     tho
    -0.25
     instant
    -0.25
    æī«æıı
    -0.25
    trad
    -0.25
     uncont
    -0.24
    èŃ
    -0.24
     Ñĩ
    -0.24
    POSITIVE LOGITS
    CAPE
    0.27
    Directed
    0.27
     Coinbase
    0.27
    psc
    0.27
    ounder
    0.26
    licable
    0.26
    ozy
    0.26
    isher
    0.25
    icles
    0.25
    嵯
    0.24
    Activations Density 0.059%

    No Known Activations