© 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. 31-TRANSCODER-HP
    7. 18567
    Prev
    Next
    INDEX
    Explanations

    timeline timing

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     left
    -0.30
    å·¦
    -0.29
    CTIONS
    -0.28
    æ°´ä¸Ń
    -0.27
    unting
    -0.26
    å·¦åı³
    -0.26
    SCI
    -0.26
    (where
    -0.26
     Left
    -0.25
    [value
    -0.25
    POSITIVE LOGITS
    chart
    0.29
     coincidence
    0.28
    illis
    0.28
    çĶ¨å¾®ä¿¡
    0.28
    esch
    0.27
    quotes
    0.26
     sheet
    0.25
    board
    0.25
     resets
    0.25
    Reset
    0.25
    Activations Density 0.819%

    No Known Activations