© 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. 20-TRANSCODER-HP
    7. 107420
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    idor
    -0.27
    endon
    -0.26
    SSI
    -0.26
     superclass
    -0.26
    (done
    -0.26
     generous
    -0.26
    åī¥
    -0.26
     göz
    -0.24
    å»ĸ
    -0.23
    readonly
    -0.23
    POSITIVE LOGITS
     offs
    0.27
     qu
    0.26
    æľ¬ç§ij
    0.24
    uant
    0.23
    éģı
    0.23
     Fal
    0.23
    æŁ¥çľĭ详æĥħ
    0.23
     Parallel
    0.23
    壮
    0.23
    éĺ¶æ®µ
    0.23
    Activations Density 17.986%

    No Known Activations

    This feature has no known activations.