© 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. 95670
    Prev
    Next
    INDEX
    Explanations

    sperm

    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
    king
    -0.27
     Transcript
    -0.27
    eda
    -0.26
     rm
    -0.26
     ju
    -0.26
     st
    -0.26
    ked
    -0.25
    -frequency
    -0.25
     transcript
    -0.25
    romium
    -0.25
    POSITIVE LOGITS
    èĥ¶åĽĬ
    0.28
    UserCode
    0.27
    utive
    0.27
     Esk
    0.26
    veloper
    0.26
    åijĬ
    0.25
    书éĻ¢
    0.25
     Sue
    0.24
    èµ°è¿Ľ
    0.24
    åŃķæľŁ
    0.24
    Activations Density 0.883%

    No Known Activations