© 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. 20-TRANSCODER-HP
    7. 64267
    Prev
    Next
    INDEX
    Explanations

    consent and permission

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    orz
    -0.31
    .Alignment
    -0.30
    示èĮĥåĮº
    -0.29
    ORS
    -0.27
    prite
    -0.26
    reau
    -0.26
    æĸĩçĮ®
    -0.25
    lÃŃ
    -0.24
    elu
    -0.24
    ningar
    -0.24
    POSITIVE LOGITS
    åŃļ
    0.32
    æĦıè§ģåıįé¦Ī
    0.28
    æĹ¥æ¶Īæģ¯
    0.27
    åĹij
    0.26
    æĹ¥æĬ¥
    0.25
     understandably
    0.25
    _hw
    0.25
     reacted
    0.24
    ivo
    0.24
    ynam
    0.24
    Activations Density 12.291%

    No Known Activations