© 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. 108780
    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
    ihn
    -0.28
    enc
    -0.26
    (ele
    -0.26
     URLRequest
    -0.26
    leta
    -0.25
    ık
    -0.25
     prod
    -0.24
     полов
    -0.24
    à¤Ī
    -0.24
     elephant
    -0.24
    POSITIVE LOGITS
     Pure
    0.26
    wner
    0.26
    Pure
    0.25
    ç½ij绾ç§ijæĬĢ
    0.24
    说äºĨ
    0.24
    linger
    0.24
    Parents
    0.24
    ÑĢаÑģÑĤ
    0.24
    ider
    0.23
    èįı
    0.23
    Activations Density 1.965%

    No Known Activations

    This feature has no known activations.