© 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. 8963
    Prev
    Next
    INDEX
    Explanations

    programming 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
    ä»Ģä¹Īäºĭæĥħ
    -0.31
    ä»Ģä¹Īäºĭ
    -0.29
    anti
    -0.28
    cq
    -0.27
    cord
    -0.26
    overflow
    -0.26
    clock
    -0.26
    odom
    -0.25
    gable
    -0.25
    agua
    -0.24
    POSITIVE LOGITS
    strate
    0.26
    illon
    0.25
     hass
    0.24
    THREAD
    0.24
    æĽ¹
    0.24
     Realm
    0.23
    å·¥å§Ķ
    0.23
     tö
    0.23
    osit
    0.23
    liers
    0.23
    Activations Density 0.039%

    No Known Activations