© 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. 109027
    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
    unks
    -0.30
    è§ģè¿ĩ
    -0.28
    ize
    -0.27
    []>(
    -0.26
    åħ±åĪĽ
    -0.25
    èĬĹ
    -0.24
    pace
    -0.24
    éĢĢ
    -0.24
    æĢ»ä¹ĭ
    -0.24
    (embed
    -0.24
    POSITIVE LOGITS
    thag
    0.29
    Letters
    0.29
     letters
    0.28
    imuth
    0.27
     Letters
    0.27
    æī¼
    0.26
    astically
    0.26
     sideline
    0.25
    طب
    0.25
     grazing
    0.25
    Activations Density 0.062%

    No Known Activations

    This feature has no known activations.