© 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. 40427
    Prev
    Next
    INDEX
    Explanations

    code, algebra, strings

    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
    b
    -0.34
    alone
    -0.31
    缺åı£
    -0.29
    agonal
    -0.28
    avel
    -0.27
    asta
    -0.27
    oul
    -0.26
    anto
    -0.26
    åľŁåľ°
    -0.26
     nale
    -0.26
    POSITIVE LOGITS
    vers
    0.31
    nect
    0.31
    mers
    0.30
    mach
    0.29
    bers
    0.29
    bing
    0.28
    ming
    0.28
     bv
    0.28
    mand
    0.27
     opr
    0.27
    Activations Density 2.075%

    No Known Activations