© 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. 33-TRANSCODER-HP
    7. 53283
    Prev
    Next
    INDEX
    Explanations

    Conservative political organizations

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     READ
    -0.28
    åĽ´çĿĢ
    -0.27
    ä»ĵ
    -0.27
    ÑĭÑĪ
    -0.25
    为æĤ¨
    -0.25
     geared
    -0.24
    éĺħ
    -0.24
    isu
    -0.24
     CPC
    -0.24
     reads
    -0.24
    POSITIVE LOGITS
    amic
    0.26
     población
    0.25
    亮
    0.25
    uts
    0.25
    éĤ»
    0.25
    wy
    0.24
    oui
    0.24
     Turns
    0.24
    KIT
    0.24
    ÙĪتر
    0.23
    Activations Density 0.001%

    No Known Activations