© 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. 54293
    Prev
    Next
    INDEX
    Explanations

    dance and music

    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
    nts
    -0.27
    conds
    -0.27
    延éķ¿
    -0.27
    bout
    -0.26
    æ±Ń
    -0.26
    æ´»çĿĢ
    -0.26
    å®ļæŶ
    -0.25
    serter
    -0.25
    ÅĤaw
    -0.25
     hott
    -0.24
    POSITIVE LOGITS
    ض
    0.27
    (super
    0.26
    esen
    0.25
    ega
    0.24
    å¼¹
    0.24
     culture
    0.24
    çļĦè¿ĩç¨ĭ
    0.24
    缴è¾ĸ
    0.24
    æĸĩåĮĸ
    0.24
    (Collection
    0.24
    Activations Density 0.020%

    No Known Activations