© 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. 4-TRANSCODER-HP
    7. 21
    Prev
    Next
    INDEX
    Explanations

    would

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_4.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.4.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    å°¸
    -0.28
    ampp
    -0.27
     poisoned
    -0.24
    çĹŀ
    -0.24
    uries
    -0.24
     sắc
    -0.24
    amp
    -0.24
    etheless
    -0.23
     bully
    -0.23
    èµĦ
    -0.23
    POSITIVE LOGITS
    ness
    0.28
    away
    0.26
    ãģĵãģ¨
    0.24
    nt
    0.24
    -known
    0.24
    大éŨ
    0.24
    è¿Ļ件
    0.24
    -initial
    0.23
    inspace
    0.23
     ditch
    0.23
    Activations Density 0.106%

    No Known Activations