© 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. 31-TRANSCODER-HP
    7. 18154
    Prev
    Next
    INDEX
    Explanations

    fair

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     equally
    -0.26
    inus
    -0.26
    itur
    -0.26
     Steel
    -0.25
    inant
    -0.25
    ç±³åħ°
    -0.24
    _macros
    -0.24
    thon
    -0.24
     MAD
    -0.24
    sprintf
    -0.24
    POSITIVE LOGITS
    athi
    0.32
    cas
    0.27
    apse
    0.25
    èĩ§
    0.25
    æłĵ
    0.25
    á»įng
    0.25
    oment
    0.24
    anonymous
    0.24
    aset
    0.24
    èĤĭ
    0.24
    Activations Density 0.037%

    No Known Activations