© 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. 30-TRANSCODER-HP
    7. 97111
    Prev
    Next
    INDEX
    Explanations

    wild

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     Silence
    -0.27
    :class
    -0.26
     rid
    -0.26
     rod
    -0.26
    odont
    -0.24
     sil
    -0.24
     silence
    -0.24
    eri
    -0.24
    eren
    -0.24
    ardon
    -0.24
    POSITIVE LOGITS
    ongyang
    0.29
    iant
    0.26
     pharmacies
    0.26
     navbar
    0.25
    compose
    0.25
    ynn
    0.25
     tế
    0.25
    çĶµåı°
    0.25
    uelles
    0.24
     Tablets
    0.24
    Activations Density 0.783%

    No Known Activations