© 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. 20-TRANSCODER-HP
    7. 108156
    Prev
    Next
    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ä¹°äºĨ
    -0.25
    éļı
    -0.24
    usc
    -0.24
     fiss
    -0.24
    oss
    -0.24
    =t
    -0.23
    éļ¨
    -0.23
    èĭĽ
    -0.23
    $$$$
    -0.23
    ãĤģ
    -0.23
    POSITIVE LOGITS
    Mel
    0.26
     POLITICO
    0.26
     Mel
    0.25
    uide
    0.25
    å®ŀéĻħè¡ĮåĬ¨
    0.24
    大家åĪĨ享
    0.24
    轨éģĵ
    0.24
     )->
    0.23
    檬
    0.23
    tools
    0.23
    Activations Density 1.582%

    No Known Activations