© 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. 53575
    Prev
    Next
    INDEX
    Explanations

    professions

    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
    ç§ijæĬĢæĪIJæŀľ
    -0.29
     Koreans
    -0.29
    hots
    -0.27
    ç½ijéĢļ
    -0.26
     bahwa
    -0.26
     Damen
    -0.25
    ç«łèĬĤ
    -0.25
    èIJ½
    -0.24
    缸éĤ»
    -0.24
     kto
    -0.24
    POSITIVE LOGITS
     and
    0.45
     special
    0.43
     best
    0.37
     active
    0.36
    ,
    0.34
     Special
    0.33
    special
    0.33
    çī¹æ®Ĭ
    0.33
    Special
    0.32
     turned
    0.32
    Activations Density 0.143%

    No Known Activations