© 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. 23-TRANSCODER-HP
    7. 25453
    Prev
    Next
    INDEX
    Explanations

    9

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_23.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.23.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    ding
    -0.30
    OnInit
    -0.28
    лоÑĤ
    -0.27
    reeze
    -0.27
     Dresden
    -0.27
    æľĢåĪĿçļĦ
    -0.26
     Matters
    -0.25
    ngo
    -0.25
    æľĢåĪĿ
    -0.25
     Structure
    -0.25
    POSITIVE LOGITS
    oland
    0.29
    untu
    0.28
     slic
    0.27
    eland
    0.27
    åĪ¤
    0.26
    cob
    0.25
    contracts
    0.25
    裸
    0.25
    lige
    0.25
    uh
    0.24
    Activations Density 0.281%

    No Known Activations