© 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. 25-TRANSCODER-HP
    7. 109359
    Prev
    Next
    INDEX
    Explanations

    can

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    çݰ代
    -0.27
     pairing
    -0.27
     modern
    -0.27
    modern
    -0.26
     acct
    -0.26
     Modern
    -0.25
    Modern
    -0.25
     moderne
    -0.25
    ...]
    -0.25
     Abilities
    -0.25
    POSITIVE LOGITS
    æķijåij½
    0.28
     Altın
    0.28
    çŁĽçĽ¾
    0.26
    alo
    0.26
     responsibility
    0.25
    æĿ¡è¯Ħ论
    0.25
    è¯ĹæĦı
    0.24
    mites
    0.24
    è¿´
    0.24
    gré
    0.24
    Activations Density 0.936%

    No Known Activations