© 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. 108826
    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
     Var
    -0.27
     Quart
    -0.25
    éĢ®
    -0.25
    Ä©
    -0.25
    à¸Ļำ
    -0.25
    cus
    -0.25
    æĮĩå¼ķ
    -0.24
    arak
    -0.24
    sted
    -0.24
     desn
    -0.24
    POSITIVE LOGITS
    anguage
    0.28
     Cly
    0.26
    [ix
    0.26
    çĶµè§Ĩ
    0.25
    еÑĩно
    0.25
    mentation
    0.25
    onents
    0.25
    ulaire
    0.24
    严åİīæīĵåĩ»
    0.24
    à¸Ħาร
    0.24
    Activations Density 0.006%

    No Known Activations

    This feature has no known activations.