© 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. 95141
    Prev
    Next
    INDEX
    Explanations

    long

    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
     genotype
    -0.26
     energies
    -0.26
    ENA
    -0.25
     mutable
    -0.25
     mailing
    -0.24
     mesa
    -0.24
    nergy
    -0.24
    elho
    -0.23
    enade
    -0.23
    alc
    -0.22
    POSITIVE LOGITS
    ers
    0.40
    erset
    0.34
    容
    0.30
    ä¿Ł
    0.28
    已被
    0.28
    容å¿į
    0.28
    常çĶ¨
    0.28
    æĥ¯
    0.28
    æłħ
    0.27
    é¦
    0.26
    Activations Density 0.007%

    No Known Activations