© 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. 97861
    Prev
    Next
    INDEX
    Explanations

    code, math, census data

    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
    éĽªå±±
    -0.27
    lik
    -0.26
     unc
    -0.25
     coveted
    -0.25
    æĿł
    -0.24
     lik
    -0.23
     and
    -0.23
     commercial
    -0.23
     criteria
    -0.23
    çĺª
    -0.23
    POSITIVE LOGITS
    tern
    0.30
    æ³Ľ
    0.28
    aroo
    0.28
    ote
    0.27
    误
    0.27
     RTL
    0.25
    '};↵
    0.25
    DMIN
    0.24
    pun
    0.24
    rror
    0.24
    Activations Density 0.001%

    No Known Activations