© 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. 89908
    Prev
    Next
    INDEX
    Explanations

    man

    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
    |string
    -0.27
     Twe
    -0.27
     compuls
    -0.27
    ModelState
    -0.25
    ä¸įåħį
    -0.25
    ORIGINAL
    -0.24
    好å¥ĩå¿ĥ
    -0.24
     fatally
    -0.24
    æĺ¯åı¯ä»¥
    -0.24
    \TestCase
    -0.23
    POSITIVE LOGITS
     recognize
    0.26
    ensis
    0.26
    loy
    0.26
    aroo
    0.25
    stein
    0.25
    ä¸įåĩĨ
    0.24
    çģ«
    0.24
     hu
    0.24
    éĥ½çŁ¥éģĵ
    0.24
    ixin
    0.24
    Activations Density 3.544%

    No Known Activations