© 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. 87327
    Prev
    Next
    INDEX
    Explanations

    Temple

    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
    éĤ¸
    -0.25
    æĶ¾åĪ°
    -0.25
    undo
    -0.23
    éĩįè¦ģ
    -0.23
     Bá
    -0.23
     Mim
    -0.23
    注æĦı
    -0.23
    patch
    -0.22
    pring
    -0.22
    éĩįçĤ¹
    -0.22
    POSITIVE LOGITS
    istry
    0.30
    perature
    0.27
    assage
    0.27
     Tues
    0.24
    orary
    0.24
    ä¸ŀ
    0.24
    icipants
    0.24
     Juda
    0.24
    oxic
    0.24
    æ¼Ĥ
    0.24
    Activations Density 0.234%

    No Known Activations