© 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. 25-TRANSCODER-HP
    7. 109315
    Prev
    Next
    INDEX
    Explanations

    scientific papers

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    æĤ£
    -0.28
    :add
    -0.27
     repeated
    -0.27
    edio
    -0.25
    许å¤ļ人
    -0.25
    çĶ·æĢ§
    -0.25
     Woods
    -0.24
     applied
    -0.24
     bad
    -0.24
     weekly
    -0.24
    POSITIVE LOGITS
    rez
    0.27
    riad
    0.24
     gradu
    0.24
    缸åħ³ä¿¡æģ¯
    0.24
    forge
    0.23
    rics
    0.23
    @class
    0.23
     cancell
    0.23
     Rise
    0.23
    .IContainer
    0.23
    Activations Density 0.890%

    No Known Activations