© 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. 54061
    Prev
    Next
    INDEX
    Explanations

    for or of

    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
     ~
    -0.38
    (~
    -0.35
     ~~
    -0.34
    ~
    -0.33
    å®ĥ
    -0.32
     ~(
    -0.32
    ~~
    -0.31
    å¾Ħ
    -0.29
     [~,
    -0.28
     (~
    -0.28
    POSITIVE LOGITS
     eleven
    0.37
     thirteen
    0.32
     one
    0.32
     seven
    0.32
     three
    0.31
     six
    0.31
     fourteen
    0.31
     Eleven
    0.30
     thirty
    0.30
     twelve
    0.30
    Activations Density 0.022%

    No Known Activations