© 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. 33-TRANSCODER-HP
    7. 54694
    Prev
    Next
    INDEX
    Explanations

    Technical/medical context

    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
    Negative Logits
    ACLE
    -0.26
    _increment
    -0.26
    acha
    -0.26
    èĿĩ
    -0.25
    olated
    -0.25
    acs
    -0.24
    aky
    -0.24
    éĴĹ
    -0.24
    éķĽ
    -0.24
    éĢĬ
    -0.24
    POSITIVE LOGITS
    upe
    0.28
    åĬ¡
    0.27
     Cyc
    0.27
    èĴĤ
    0.26
    ä¸Ĭä¸ĭ
    0.25
    rough
    0.25
    æľ¬è½®
    0.25
    utions
    0.25
    èļķ
    0.24
    kick
    0.24
    Activations Density 0.003%

    No Known Activations