© 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. 150197
    Prev
    Next
    INDEX
    Explanations

    Words expressing opinions

    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
     indirect
    -0.25
    -resolution
    -0.23
    agi
    -0.23
    åĨ·éĿĻ
    -0.23
    -minus
    -0.23
    alla
    -0.23
     resolution
    -0.23
    æģ°å½ĵ
    -0.23
    解åĨ³äºĨ
    -0.23
    -settings
    -0.23
    POSITIVE LOGITS
     reportedly
    0.31
     supposedly
    0.30
     routinely
    0.27
     oft
    0.26
    常æĢģåĮĸ
    0.25
     allegedly
    0.25
    apesh
    0.25
     stuck
    0.25
    Desk
    0.24
     originally
    0.23
    Activations Density 0.892%

    No Known Activations