© 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. 112281
    Prev
    Next
    INDEX
    Explanations

    Knowledge, uncertainty

    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
    çĥŁ
    -0.28
    ws
    -0.27
     Smoke
    -0.26
    ä¼ij
    -0.26
     bulk
    -0.25
    Ïĭ
    -0.25
    :',↵
    -0.25
     negot
    -0.25
    Ñĺ
    -0.25
    sd
    -0.25
    POSITIVE LOGITS
    asca
    0.26
    /reset
    0.26
     /↵↵
    0.26
     '-'
    0.26
    kräfte
    0.25
    -serving
    0.25
    .gstatic
    0.25
    åį°
    0.24
     affine
    0.24
    ..'
    0.24
    Activations Density 0.021%

    No Known Activations