© 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. 31-TRANSCODER-HP
    7. 19794
    Prev
    Next
    INDEX
    Explanations

    Ay

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    hammer
    -0.30
     char
    -0.29
    mue
    -0.29
    èĦ¯
    -0.28
    å¦Ĥæŀľæ²¡æľī
    -0.28
     fundamental
    -0.27
    acock
    -0.26
    æł¹æľ¬å°±ä¸į
    -0.26
    char
    -0.25
    NavController
    -0.25
    POSITIVE LOGITS
    cent
    0.26
    ivities
    0.26
    ieten
    0.25
    åĬŁè¯¾
    0.25
    seau
    0.25
     prune
    0.24
    ç³Ł
    0.24
     blink
    0.24
    ivi
    0.24
    .goal
    0.24
    Activations Density 0.004%

    No Known Activations