© 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. 30-TRANSCODER-HP
    7. 95612
    Prev
    Next
    INDEX
    Explanations

    Lift

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    oard
    -0.27
    éĽĨæĪIJ
    -0.26
    kich
    -0.26
    è¯įæĿ¡
    -0.26
    $ar
    -0.25
     chatter
    -0.25
    æł·çļĦ
    -0.25
    æĩĤäºĭ
    -0.25
    ä½ľåĵģåĨħ容
    -0.24
     semen
    -0.24
    POSITIVE LOGITS
     lift
    0.47
    lift
    0.45
     elev
    0.45
     Elev
    0.40
     lifts
    0.38
     Lift
    0.35
     lifted
    0.33
    _lift
    0.32
    .lift
    0.32
     elevator
    0.31
    Activations Density 0.887%

    No Known Activations