© 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. 32-TRANSCODER-HP
    7. 44199
    Prev
    Next
    INDEX
    Explanations

    build

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    romatic
    -0.29
    éĺģ
    -0.28
    arden
    -0.28
    義
    -0.26
     hors
    -0.24
    egrated
    -0.24
    åĪĩåī²
    -0.24
    鼶ç¢İ
    -0.23
    ä¹ī
    -0.23
    ard
    -0.23
    POSITIVE LOGITS
    -produced
    0.36
     builds
    0.35
    -built
    0.35
     build
    0.32
     Build
    0.32
     Builds
    0.32
     buildup
    0.31
    _Build
    0.30
     built
    0.30
     levels
    0.30
    Activations Density 0.000%

    No Known Activations