© 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. 41902
    Prev
    Next
    INDEX
    Explanations

    lists of advice/things to know

    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
     distribute
    -0.35
     protect
    -0.33
     connect
    -0.33
     express
    -0.32
     extend
    -0.32
     represent
    -0.31
     deploy
    -0.30
     arrange
    -0.30
     expand
    -0.30
     collect
    -0.29
    POSITIVE LOGITS
     GetCurrent
    0.31
    ừa
    0.28
    :Get
    0.28
    åĴĮæĬĢæľ¯
    0.27
     GetValue
    0.27
     literally
    0.27
    (Get
    0.26
    _passed
    0.26
     GetById
    0.26
    GetObject
    0.26
    Activations Density 0.005%

    No Known Activations