© 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. 33-TRANSCODER-HP
    7. 53782
    Prev
    Next
    INDEX
    Explanations

    technical and political text

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Rae
    -0.26
    ä¿ĿåĪ©
    -0.26
    æ¤Ĵ
    -0.25
    무
    -0.25
    ç¥ĸ
    -0.24
    ãģĮå§ĭ
    -0.24
    RAP
    -0.23
    ullah
    -0.23
     Ramos
    -0.23
    .$.
    -0.23
    POSITIVE LOGITS
    å¼Ģåıij建设
    0.26
    енно
    0.26
    pressions
    0.25
    EXTERN
    0.25
    ouched
    0.24
    建设工ç¨ĭ
    0.24
    hores
    0.24
     Mold
    0.23
    åľľ
    0.23
    acias
    0.23
    Activations Density 0.002%

    No Known Activations