© 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. 98121
    Prev
    Next
    INDEX
    Explanations

    Actions/performing actions

    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
    asz
    -0.30
    èµ°ä¸Ĭ
    -0.28
    峡
    -0.26
    è§Ĥ
    -0.25
    åįģåĽĽäºĶ
    -0.25
    çĻ½äºij
    -0.25
    æĮ¯ãĤĬ
    -0.24
    ubes
    -0.23
    -controller
    -0.23
     as
    -0.23
    POSITIVE LOGITS
    ported
    0.26
     [~,
    0.25
     FAG
    0.25
    Touch
    0.24
    iets
    0.24
    owoÅĽci
    0.23
    çļĦèĦļæŃ¥
    0.23
    nı
    0.23
    ISED
    0.23
     dej
    0.23
    Activations Density 0.009%

    No Known Activations