© 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. 53435
    Prev
    Next
    INDEX
    Explanations

    match

    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
     improved
    -0.33
     reduced
    -0.28
    è¿Ľ
    -0.28
    æĶ¶åĽŀ
    -0.26
     beneficiaries
    -0.25
     advantageous
    -0.25
    åıĸ
    -0.25
     stands
    -0.24
    ĥĿ
    -0.24
    Improved
    -0.24
    POSITIVE LOGITS
     matched
    0.38
    being
    0.37
    matched
    0.33
    åĮ¹éħį
    0.33
     match
    0.32
     Match
    0.32
    éħį
    0.31
     be
    0.30
    _match
    0.30
     MATCH
    0.29
    Activations Density 0.013%

    No Known Activations