© 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. 108534
    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
    ActionResult
    -0.26
    æį®
    -0.25
     subjects
    -0.24
     daily
    -0.24
    atoi
    -0.24
    èĥ½å¤Ł
    -0.24
    æ
    -0.24
     charges
    -0.23
     nightly
    -0.23
    æĵļ
    -0.23
    POSITIVE LOGITS
    åĪ¹
    0.30
     <![
    0.29
    ragen
    0.28
    IVITY
    0.27
    Witness
    0.27
    æĹħ游èµĦæºIJ
    0.27
    collapsed
    0.25
    ัย
    0.24
    åĸīåĴĻ
    0.24
    راج
    0.24
    Activations Density 1.043%

    No Known Activations

    This feature has no known activations.