© 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. 109289
    Prev
    Next
    INDEX
    Explanations

    Code and file paths

    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
    idor
    -0.31
     col
    -0.27
     comp
    -0.27
     radicals
    -0.27
     fore
    -0.27
    ably
    -0.27
     intercept
    -0.26
    ä½¥
    -0.26
     hyp
    -0.25
    IDER
    -0.24
    POSITIVE LOGITS
    认æ¸ħ
    0.27
     frais
    0.26
    baugh
    0.25
    å±ĢéĻIJ
    0.25
    åı®
    0.24
    xfe
    0.24
    ern
    0.24
    פ׳×Ļ
    0.23
     Entered
    0.23
    å²Ń
    0.23
    Activations Density 3.508%

    No Known Activations