© 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. 129512
    Prev
    Next
    INDEX
    Explanations

    Ranked lists and information

    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
    ingham
    -0.34
    etermination
    -0.28
    å¢ŀéĩı
    -0.28
    弥补
    -0.28
    åĬ¡
    -0.27
     proving
    -0.25
    æŁĦ
    -0.25
    increment
    -0.25
     determination
    -0.25
    IENT
    -0.25
    POSITIVE LOGITS
    ä¸Ģç¬ij
    0.27
     subt
    0.27
     standing
    0.25
     regular
    0.25
    çĤ¹å¤´
    0.25
    tele
    0.25
    standing
    0.24
     gord
    0.24
    iard
    0.24
    rees
    0.23
    Activations Density 2.995%

    No Known Activations