© 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. 38754
    Prev
    Next
    INDEX
    Explanations

    reported information

    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
    Negative Logits
     BBB
    -0.33
     PPP
    -0.31
    fsp
    -0.30
     Tb
    -0.30
     HH
    -0.30
    HH
    -0.30
     MMM
    -0.29
    PPP
    -0.29
     EIF
    -0.29
     EH
    -0.29
    POSITIVE LOGITS
    å½ĵä¹ĭæĹłæĦ§
    0.35
    —an
    0.32
    —a
    0.31
    è½»è½»åľ°
    0.29
    —they
    0.29
    —the
    0.29
    —it
    0.29
    ä¸įåģľåľ°
    0.28
    —not
    0.27
    è¿ŀå¿Ļ
    0.27
    Activations Density 18.935%

    No Known Activations