© 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. 96361
    Prev
    Next
    INDEX
    Explanations

    legal and criminal cases

    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
    is
    -0.28
    æ¶ķ
    -0.27
    ä¸Ģ声
    -0.26
    çīĪæĿĥå½Ĵ
    -0.25
    æ²³æµģåŁŁ
    -0.25
    åĿĬ
    -0.25
    æĤ²
    -0.24
    Reuse
    -0.24
    å¯ĨéĽĨ
    -0.24
    æĶ¯ä»ĺ
    -0.24
    POSITIVE LOGITS
    lage
    0.28
     Jens
    0.27
    åĢ¼çıŃ
    0.26
    ç¼ĵ
    0.26
     DÆ°Æ¡ng
    0.26
    _PARSE
    0.26
    pector
    0.25
    HING
    0.25
     unloaded
    0.25
     accomp
    0.24
    Activations Density 0.929%

    No Known Activations