© 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. 41193
    Prev
    Next
    INDEX
    Explanations

    Testing

    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
    Not in Any Lists

    No Comments

    Negative Logits
    ubs
    -0.32
     justice
    -0.28
     balance
    -0.27
    usive
    -0.26
     bean
    -0.26
    usi
    -0.26
    usize
    -0.26
     fac
    -0.26
    çĴľ
    -0.26
     Cuban
    -0.26
    POSITIVE LOGITS
    weg
    0.30
    çĭ¡
    0.28
    er
    0.27
    æľ¯è¯Ń
    0.27
    æ³Ħ
    0.26
    ÅĽl
    0.25
    èĦ±
    0.25
    word
    0.25
    诡
    0.24
    Disclosure
    0.24
    Activations Density 0.001%

    No Known Activations