© 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. 25-TRANSCODER-HP
    7. 109141
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Sizes
    -0.28
     ordained
    -0.28
    Sizes
    -0.28
    ighting
    -0.27
    sizes
    -0.26
    ä¹ĸ
    -0.26
    å¾µ
    -0.25
     RECE
    -0.25
    GLenum
    -0.25
    ä¸įç»ı
    -0.24
    POSITIVE LOGITS
    注
    0.28
    mium
    0.28
    ttl
    0.27
    æī¿è®¤
    0.26
    ãģ»ãģ©
    0.25
    ões
    0.24
     drill
    0.24
     boarded
    0.23
    å¡Į
    0.23
     discount
    0.23
    Activations Density 0.112%

    No Known Activations

    This feature has no known activations.