© 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. 94964
    Prev
    Next
    INDEX
    Explanations

    El

    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
    Elf
    -0.33
    Self
    -0.31
     self
    -0.30
    self
    -0.30
     Self
    -0.26
     catching
    -0.25
    oring
    -0.25
    [self
    -0.25
    elf
    -0.25
    /self
    -0.24
    POSITIVE LOGITS
    çIJĨ解åĴĮ
    0.28
    atic
    0.26
    ight
    0.26
    çĹĺ
    0.26
    å¨ĺ
    0.26
    mtree
    0.25
    tingham
    0.25
    .isActive
    0.25
    æij©
    0.25
    isSelected
    0.25
    Activations Density 0.044%

    No Known Activations