© 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. 96931
    Prev
    Next
    INDEX
    Explanations

    uring

    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
     ph
    -0.31
     going
    -0.27
    çķĻå®Ī
    -0.27
    Starting
    -0.27
    Bundle
    -0.26
    åł¡
    -0.26
    LoggedIn
    -0.25
     Going
    -0.25
     Starting
    -0.25
    åĨįåİ»
    -0.25
    POSITIVE LOGITS
    rog
    0.27
    matchCondition
    0.27
    irl
    0.26
    ä¸Ńè¶ħ
    0.26
    hang
    0.26
    bens
    0.26
    iteur
    0.26
    对æīĭ
    0.25
    ipl
    0.24
    åħ¨æĸĩ
    0.24
    Activations Density 0.830%

    No Known Activations