© 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. 130085
    Prev
    Next
    INDEX
    Explanations

    Quotes and parenthetical asides

    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
    lfw
    -0.29
    ä½ľä¸º
    -0.27
    =find
    -0.26
     als
    -0.26
    fall
    -0.25
    eton
    -0.25
     jako
    -0.25
    èŃ
    -0.24
    (reinterpret
    -0.24
    èħ®
    -0.23
    POSITIVE LOGITS
    äºĮåŃĹ
    0.36
     keyword
    0.34
     adjective
    0.29
    åħ³éĶ®è¯į
    0.28
     word
    0.28
    rey
    0.28
     palavra
    0.27
    (keyword
    0.27
    /crypto
    0.27
    -word
    0.26
    Activations Density 43.462%

    No Known Activations