© 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. 22-TRANSCODER-HP
    7. 87566
    Prev
    Next
    INDEX
    Explanations

    the

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    éķij
    -0.28
    ([],
    -0.27
    vor
    -0.26
    ез
    -0.25
    หà¸Ļ
    -0.25
    åѤ
    -0.25
     side
    -0.25
     inv
    -0.24
    é£İéĻ©ç®¡çIJĨ
    -0.24
     seria
    -0.24
    POSITIVE LOGITS
     University
    0.27
    Natural
    0.26
    大桥
    0.25
     Natural
    0.25
    apse
    0.25
     Hubbard
    0.25
    çļĦçľ¼
    0.24
     Complex
    0.24
    çļĦæĶ¯æĮģ
    0.24
    çļĦ妻åŃIJ
    0.24
    Activations Density 0.027%

    No Known Activations