© 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. 18907
    Prev
    Next
    INDEX
    Explanations

    start

    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
    FX
    -0.26
    bots
    -0.26
    _ready
    -0.25
     Bot
    -0.25
     newsp
    -0.25
    IDI
    -0.24
    зал
    -0.24
    è¨ĢãģĦ
    -0.24
    浡
    -0.24
     setResult
    -0.23
    POSITIVE LOGITS
     between
    0.39
     up
    0.33
    between
    0.32
     about
    0.30
     Between
    0.30
    å¹´åIJİ
    0.29
    约
    0.28
    ä¹ĭéĹ´çļĦ
    0.28
    up
    0.28
    Between
    0.28
    Activations Density 0.105%

    No Known Activations