© 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. 18-TRANSCODER-HP
    7. 88875
    Prev
    Next
    INDEX
    Explanations

    syndrome

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_18.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.18.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    å®ĥæĺ¯
    -0.25
    ä½łæĺ¯
    -0.24
     scor
    -0.24
    rell
    -0.24
     ['$
    -0.23
    çĹī
    -0.23
    ',"
    -0.23
    .SDK
    -0.23
    ç±½
    -0.23
     sæ
    -0.22
    POSITIVE LOGITS
    vit
    0.27
     vacancy
    0.25
    isko
    0.25
    yard
    0.24
    dle
    0.24
    lage
    0.24
     eve
    0.24
    pieces
    0.24
     papers
    0.23
    (relative
    0.23
    Activations Density 11.974%

    No Known Activations