© 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. 19-TRANSCODER-HP
    7. 8536
    Prev
    Next
    INDEX
    Explanations

    survive

    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
    åĵģä½į
    -0.27
    elim
    -0.25
    æ¯Ķä¾ĭ
    -0.24
     Jonas
    -0.24
    以åīį
    -0.24
    iced
    -0.23
    otation
    -0.23
     TáºŃp
    -0.23
    edor
    -0.22
    sk
    -0.22
    POSITIVE LOGITS
    å¼Ģå¥ĸ
    0.26
    èĭ¦èĭ¦
    0.26
     cú
    0.25
     invalidated
    0.24
     aggreg
    0.24
    èľ·
    0.24
    /trans
    0.24
    å½ĺ
    0.23
    CTest
    0.23
    /read
    0.23
    Activations Density 0.464%

    No Known Activations