© 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. 53145
    Prev
    Next
    INDEX
    Explanations

    mathematical/scientific notation

    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
     âĪ
    -0.41
    sim
    -0.29
    adin
    -0.28
     (âĪ
    -0.26
    Sim
    -0.26
     soak
    -0.25
    adal
    -0.25
    bil
    -0.24
    TRS
    -0.24
    èĬ¯
    -0.24
    POSITIVE LOGITS
     more
    0.40
    æĽ´å¤ļ
    0.35
    æĽ´å¤ļçļĦ
    0.35
    more
    0.29
    _more
    0.28
     hơn
    0.27
     More
    0.27
    More
    0.27
    建åħļ
    0.26
    ve
    0.26
    Activations Density 0.010%

    No Known Activations