© 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. 109253
    Prev
    Next
    INDEX
    Explanations

    certain

    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
    Not in Any Lists

    No Comments

    Negative Logits
    KI
    -0.25
    Gran
    -0.25
    èĺŃ
    -0.25
    é²ij
    -0.25
    pare
    -0.24
     the
    -0.23
    åħ¶ä½Ļ
    -0.23
    ä¾ĺ
    -0.23
    ÙģصÙĦ
    -0.23
     both
    -0.23
    POSITIVE LOGITS
    otions
    0.28
    æľīè¿ĩ
    0.28
    alyzed
    0.27
     types
    0.26
     trib
    0.25
    gien
    0.25
    æĮģ
    0.24
    ieurs
    0.24
    onga
    0.24
    ies
    0.24
    Activations Density 0.265%

    No Known Activations