© 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. 35-TRANSCODER-HP
    7. 40417
    Prev
    Next
    INDEX
    Explanations

    say "r" words

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_35.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.35.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    æĢ§åĴĮ
    -0.29
    (S
    -0.27
    åŀĭä¼ģä¸ļ
    -0.27
    "].(
    -0.26
    åĮĸçļĦ
    -0.26
    æĿ¡è¯Ħ论
    -0.26
    ']]['
    -0.26
     $($
    -0.25
     setResult
    -0.25
    è¿ĩçļĦ
    -0.25
    POSITIVE LOGITS
    r
    0.43
    rif
    0.41
    rac
    0.40
    rat
    0.40
    rise
    0.38
    rain
    0.38
    rw
    0.35
    rack
    0.35
    ran
    0.35
    rate
    0.34
    Activations Density 3.112%

    No Known Activations