© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 23-TRANSCODER-HP
    7. 162997
    Prev
    Next
    INDEX
    Explanations

    Names/Abbreviations/Codes

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_23.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.23.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    etary
    -0.30
    JOR
    -0.29
    stadt
    -0.27
    adero
    -0.26
    çļĦåľ°
    -0.25
    urat
    -0.25
    ...)↵↵
    -0.25
    ...');↵
    -0.24
    estead
    -0.24
    reich
    -0.24
    POSITIVE LOGITS
    并
    0.31
    ä¸Ģ份
    0.28
    Ve
    0.28
    åĬ¿
    0.28
    Normalization
    0.27
    types
    0.26
    ç¼ĺ
    0.26
    æĶ¶
    0.25
    Term
    0.25
    alle
    0.25
    Activations Density 0.001%

    No Known Activations