© 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. 89929
    Prev
    Next
    INDEX
    Explanations

    expressing intentions

    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
    为äºĨ让
    -0.28
    åľ¨çĶŁæ´»ä¸Ń
    -0.27
    byn
    -0.27
    为äºĨ
    -0.26
    _books
    -0.26
    æĬķå½±
    -0.26
    书ä¸Ń
    -0.25
    deo
    -0.24
    books
    -0.24
     знаÑĤÑĮ
    -0.24
    POSITIVE LOGITS
     properly
    0.30
     suitability
    0.27
     valued
    0.26
     respected
    0.26
    éĢļå¾Ģ
    0.25
    ìľ¨
    0.24
    pile
    0.24
    æ«ĥ
    0.23
     dirs
    0.23
    -stack
    0.23
    Activations Density 31.906%

    No Known Activations