© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 19-TRANSCODER-HP
    7. 6034
    Prev
    Next
    INDEX
    Explanations

    emphasize

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_19.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.19.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    裹
    -0.27
    ä»Ĺ
    -0.26
     heck
    -0.26
     hell
    -0.26
    çıŃ
    -0.25
    Calling
    -0.25
    è¿ĺ
    -0.25
    订åįķ
    -0.24
    åıŃ
    -0.24
    hon
    -0.23
    POSITIVE LOGITS
    æ¸İ
    0.27
     #=>
    0.26
     tear
    0.26
    instanc
    0.24
    umes
    0.24
    umo
    0.24
     erection
    0.23
    ube
    0.23
    shed
    0.23
     оÑĢигинал
    0.23
    Activations Density 0.019%

    No Known Activations