© 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. 63942
    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
    éĽİ
    -0.29
    ANCES
    -0.25
    unda
    -0.25
    mana
    -0.25
    veau
    -0.25
    ç®´
    -0.25
    ×ķת
    -0.24
    ansom
    -0.24
    ventus
    -0.24
    liest
    -0.24
    POSITIVE LOGITS
    æ®ĭ
    0.30
    è¾¹
    0.29
    åĪĹ
    0.27
    ite
    0.26
     Nich
    0.26
    col
    0.26
    aside
    0.25
    éļIJ
    0.24
    æĬĬæīĭ
    0.24
     NH
    0.24
    Activations Density 13.228%

    No Known Activations

    This feature has no known activations.