© 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. 111890
    Prev
    Next
    INDEX
    Explanations

    punctuation

    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
     bitcoin
    -0.28
    edb
    -0.26
     HAPP
    -0.26
     tats
    -0.26
     isNaN
    -0.25
     happ
    -0.25
    uali
    -0.25
    алÑĮнÑĭе
    -0.24
    TeX
    -0.24
    ÑģÑĤвеннÑĭе
    -0.24
    POSITIVE LOGITS
    /-
    0.28
    çĶ·ç¥ŀ
    0.27
    vit
    0.26
     skirt
    0.25
    wand
    0.25
     skirts
    0.25
    (priority
    0.25
    å¸Ń
    0.24
    pat
    0.24
    çĩĥ
    0.24
    Activations Density 0.009%

    No Known Activations