© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 35-TRANSCODER-HP
    7. 39007
    Prev
    Next
    INDEX
    Explanations

    organizations and governments

    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
    xdd
    -0.29
    æijĶ
    -0.29
    bbe
    -0.28
    -eyed
    -0.27
    ixa
    -0.27
    -return
    -0.26
     freaking
    -0.26
    è·³åĩº
    -0.26
    -empty
    -0.26
     damn
    -0.26
    POSITIVE LOGITS
     Legislative
    0.30
     MBA
    0.29
     Constitution
    0.29
     Prosecutor
    0.28
     sourceMappingURL
    0.28
    Officers
    0.28
     APK
    0.28
     Minister
    0.27
     The
    0.27
     Politics
    0.27
    Activations Density 14.104%

    No Known Activations