© 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. 33-TRANSCODER-HP
    7. 54363
    Prev
    Next
    INDEX
    Explanations

    variations of "del"

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    çĿ«æ¯Ľ
    -0.25
    +W
    -0.24
    æİ´
    -0.23
    odox
    -0.23
    éĥ½ä¸įä¼ļ
    -0.23
    ItemClick
    -0.23
     Malk
    -0.23
    ÙĤب
    -0.23
    æıIJèµ·
    -0.23
    åı°åĮĹ
    -0.22
    POSITIVE LOGITS
     del
    1.59
    del
    1.28
    	del
    1.17
     Del
    1.14
    .del
    1.13
    Del
    1.12
    (del
    1.12
    _del
    1.07
     DEL
    1.00
    -del
    1.00
    Activations Density 0.443%

    No Known Activations