© 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. 30-TRANSCODER-HP
    7. 96529
    Prev
    Next
    INDEX
    Explanations

    John

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    åĪ¤
    -0.29
    ứng
    -0.28
    è®°ä½ı
    -0.25
    çĤ¹çĤ¹å¤´
    -0.25
    举
    -0.25
    治好
    -0.25
    åıij
    -0.25
    ~-
    -0.25
    )||
    -0.24
    ç»ĺ
    -0.24
    POSITIVE LOGITS
    thic
    0.29
     Elim
    0.26
     reserv
    0.25
    getic
    0.25
    ULE
    0.24
    Elim
    0.24
    hev
    0.24
    gie
    0.24
    oload
    0.23
    Scripts
    0.23
    Activations Density 0.080%

    No Known Activations