© 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. 30-TRANSCODER-HP
    7. 95211
    Prev
    Next
    INDEX
    Explanations

    der

    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
    zet
    -0.27
    vro
    -0.27
    eteor
    -0.26
     Brew
    -0.25
    è¡Ĺä¸Ĭ
    -0.25
    vt
    -0.24
     unto
    -0.24
    æĤ¨åı¯ä»¥
    -0.24
    Ts
    -0.24
    æī¹å¤į
    -0.24
    POSITIVE LOGITS
    enger
    0.33
    åįĹå±±
    0.27
    enta
    0.27
    åįĹ
    0.27
    igg
    0.26
    icus
    0.26
    icot
    0.26
    iful
    0.25
    antas
    0.25
    oundingBox
    0.25
    Activations Density 0.022%

    No Known Activations