© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 22-TRANSCODER-HP
    7. 130855
    Prev
    Next
    INDEX
    Explanations

    copyright

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     figure
    -0.26
    éri
    -0.25
     president
    -0.24
    ä¸įæħİ
    -0.24
     er
    -0.24
    åѦ龸
    -0.24
    çľĭå®Į
    -0.23
    强度
    -0.23
     compan
    -0.23
     wouldn
    -0.23
    POSITIVE LOGITS
    êµIJ
    0.25
    è¿Ļä¸ªåľ°åĽ¾
    0.24
     vertically
    0.24
    orte
    0.24
     NodeType
    0.24
     Vertical
    0.24
     ordering
    0.24
    vertical
    0.24
    xes
    0.24
    ç¨ĭå¼ı
    0.23
    Activations Density 0.005%

    No Known Activations