© 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. 22-TRANSCODER-HP
    7. 112005
    Prev
    Next
    INDEX
    Explanations

    numerals

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    èĪī
    -0.29
    hält
    -0.28
    lias
    -0.28
    çĭĢ
    -0.27
    lian
    -0.27
    icine
    -0.26
    ieber
    -0.26
    举
    -0.26
    zw
    -0.25
    خت
    -0.25
    POSITIVE LOGITS
    è¦ģä¹Ī
    0.28
     hone
    0.26
    æĻ®éĢļ人
    0.25
    _COMPARE
    0.24
    -addons
    0.24
    墩
    0.24
    nop
    0.24
     vs
    0.24
    åѤç«ĭ
    0.23
     '../../../../../
    0.23
    Activations Density 0.003%

    No Known Activations