© 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. 24-TRANSCODER-HP
    7. 150115
    Prev
    Next
    INDEX
    Explanations

    general descriptions

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_24.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.24.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ikh
    -0.26
     Place
    -0.25
    illation
    -0.25
    ifer
    -0.24
    fty
    -0.24
    Concurrency
    -0.24
    ä¸įå®Į
    -0.24
    omat
    -0.24
    omega
    -0.24
    è¢ĸ
    -0.24
    POSITIVE LOGITS
    å¿łè¯ļ
    0.31
     susceptible
    0.28
     playable
    0.26
     cleanly
    0.26
     damaging
    0.26
    åĬŀ好
    0.26
     natural
    0.26
    _owner
    0.25
    natural
    0.25
     forgiving
    0.25
    Activations Density 0.128%

    No Known Activations