© 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. 20-TRANSCODER-HP
    7. 106960
    Prev
    Next
    INDEX
    Explanations

    General text passages

    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
     moment
    -0.27
     recovery
    -0.27
     impuls
    -0.26
    帮åĬ©ä¼ģä¸ļ
    -0.26
    */↵
    -0.26
    åIJĥçļĦ
    -0.26
    issan
    -0.25
    æĶ¯çº¿
    -0.25
     Norton
    -0.25
    idia
    -0.25
    POSITIVE LOGITS
     incarnation
    0.25
    æ»Ĥ
    0.24
    ungan
    0.24
    æĪijå¸Ĥ
    0.23
    è°ĥæݧ
    0.23
    ubi
    0.23
     Gerald
    0.23
    spin
    0.23
    .atomic
    0.23
    .orig
    0.23
    Activations Density 12.607%

    No Known Activations