© 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. 22-TRANSCODER-HP
    7. 97280
    Prev
    Next
    INDEX
    Explanations

    Geology and rock formations

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    èĥ½ä½¿
    -0.28
    oho
    -0.26
     Titles
    -0.25
    ç©¿
    -0.25
     beb
    -0.25
     dep
    -0.25
     Spotlight
    -0.24
     Recognition
    -0.24
    çĦ¦
    -0.24
    å½ĵåīįä½įç½®
    -0.24
    POSITIVE LOGITS
    @stop
    0.28
     scor
    0.28
    ศาส
    0.27
    ï¼ģ↵↵↵↵
    0.26
    (SC
    0.26
    proper
    0.25
    éĶ¢
    0.25
    -dominated
    0.25
    ç»ĪæŃ¢
    0.24
    sex
    0.24
    Activations Density 0.006%

    No Known Activations