© 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. 23-TRANSCODER-HP
    7. 24725
    Prev
    Next
    INDEX
    Explanations

    in for

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_23.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.23.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    miss
    -0.28
    éģĹ
    -0.27
    erta
    -0.27
    /MPL
    -0.26
    æĺ¯ä¸Ģä½į
    -0.26
    ruise
    -0.26
    鸩
    -0.25
    uzzer
    -0.25
    çĶ±ä¸ŃåĽ½
    -0.25
    lected
    -0.25
    POSITIVE LOGITS
    andes
    0.29
    ãĢģãĢģ
    0.25
    OnUiThread
    0.25
    -Cs
    0.25
     bắt
    0.25
    irk
    0.24
    äºĭåĬ¡
    0.24
    GetName
    0.24
    getColor
    0.24
    ãĢģ
    0.24
    Activations Density 0.001%

    No Known Activations