© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 25-TRANSCODER-HP
    7. 107076
    Prev
    Next
    INDEX
    Explanations

    a

    np_max-act-logits · gemini-2.0-flash
    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
    Negative Logits
    fx
    -0.25
    带头人
    -0.25
    ائÙĨ
    -0.25
    igung
    -0.25
    ãģ¾ãģ¨
    -0.24
    ammable
    -0.24
    ç®Ĺæĺ¯
    -0.24
    æĪIJ为ä¸ŃåĽ½
    -0.24
    æĪIJ为
    -0.24
    burger
    -0.24
    POSITIVE LOGITS
    LETE
    0.31
    ifest
    0.28
    åħ¶æ¬¡
    0.27
    æĭĪ
    0.26
    hi
    0.26
    ORD
    0.26
    LEVEL
    0.25
    lete
    0.25
     ride
    0.25
     nature
    0.24
    Activations Density 1.088%

    No Known Activations