© 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. 22-TRANSCODER-HP
    7. 129518
    Prev
    Next
    INDEX
    Explanations

    before

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    å°Ĭ严
    -0.28
    å°±è¿ij
    -0.26
    ä¸įè¦ģ太
    -0.26
    izen
    -0.26
    EO
    -0.25
    haps
    -0.25
     grâce
    -0.25
    空åīį
    -0.25
    exclude
    -0.24
     hadn
    -0.24
    POSITIVE LOGITS
     continuation
    0.30
    继ç»Ń
    0.30
     multiples
    0.27
    继ç»Ńä¿ĿæĮģ
    0.27
    =ax
    0.27
     continue
    0.26
     resumes
    0.26
     Burnett
    0.25
     coincidence
    0.25
    hoff
    0.25
    Activations Density 0.011%

    No Known Activations