© 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. 54132
    Prev
    Next
    INDEX
    Explanations

    hurricanes

    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
    纪
    -0.27
    utf
    -0.26
    ination
    -0.26
    ainer
    -0.26
     monoc
    -0.25
    åºı
    -0.24
     metab
    -0.24
    Sil
    -0.24
     Alleg
    -0.23
    enÃŃ
    -0.23
    POSITIVE LOGITS
    odus
    0.28
    鸾
    0.27
    ót
    0.27
    alam
    0.26
    oda
    0.26
    gaard
    0.26
    aled
    0.24
    ãĥĪãĥ³
    0.24
    ears
    0.24
    èģ½
    0.24
    Activations Density 0.002%

    No Known Activations