© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 33-TRANSCODER-HP
    7. 52636
    Prev
    Next
    INDEX
    Explanations

    occupy/up

    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
    Negative Logits
    oreach
    -0.28
    èĬ®
    -0.28
    å½ĵéĢī
    -0.26
    !.↵↵
    -0.25
    tras
    -0.25
     ProductService
    -0.25
     alertController
    -0.24
    actic
    -0.24
    ,Th
    -0.24
    ñas
    -0.24
    POSITIVE LOGITS
     room
    0.40
     rooms
    0.39
     far
    0.37
    room
    0.33
    rooms
    0.32
    ä¸įå¤ļ
    0.32
     ground
    0.29
     part
    0.29
     zeroes
    0.28
     halfway
    0.27
    Activations Density 0.022%

    No Known Activations