© 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. 88204
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
    åŁº
    -0.30
    _baseline
    -0.28
     bas
    -0.28
    -offset
    -0.26
    æķı
    -0.25
    ç½ijç«Ļé¦ĸ页
    -0.25
    å·®
    -0.25
     pinch
    -0.25
    -kit
    -0.25
    æ®ĭ
    -0.24
    POSITIVE LOGITS
    FC
    0.26
    rror
    0.25
    沨
    0.23
    thumbs
    0.23
    iface
    0.23
    æĭĴ
    0.22
    (fc
    0.22
    çĺ´
    0.22
    ç¾ij
    0.22
    æªĢ
    0.22
    Activations Density 0.002%

    No Known Activations

    This feature has no known activations.