© 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. 32-TRANSCODER-HP
    7. 42492
    Prev
    Next
    INDEX
    Explanations

    of

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    amps
    -0.28
    -tax
    -0.27
    aved
    -0.27
     functional
    -0.26
    å¥Ķ
    -0.25
     dom
    -0.25
    itant
    -0.25
     meets
    -0.24
    iant
    -0.24
    .addListener
    -0.24
    POSITIVE LOGITS
    IDL
    0.28
    æĽ´æ·±
    0.27
    publication
    0.26
    DH
    0.26
    åĽ½æľīä¼ģä¸ļ
    0.26
    æħĭ
    0.25
    liches
    0.25
    å¯Ł
    0.25
    eneg
    0.25
    cken
    0.25
    Activations Density 0.003%

    No Known Activations