© 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. 35-TRANSCODER-HP
    7. 39120
    Prev
    Next
    INDEX
    Explanations

    Scandinavian languages

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_35.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.35.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    kich
    -0.28
    athers
    -0.28
    èIJ½ãģ¡
    -0.27
    cion
    -0.27
    åĬłæ·±
    -0.26
    åĽłä¸ºæĪij们
    -0.26
    ä¾Ĺ
    -0.26
     Zion
    -0.26
    scription
    -0.26
    cence
    -0.26
    POSITIVE LOGITS
    j
    0.31
     Traditional
    0.31
    å
    0.30
    je
    0.29
    jer
    0.29
     traditional
    0.28
    Traditional
    0.27
    J
    0.27
    enn
    0.26
    åŃIJéĩĮ
    0.26
    Activations Density 6.907%

    No Known Activations