© 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. 95323
    Prev
    Next
    INDEX
    Explanations

    complete

    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
    æī¬
    -0.31
    CCC
    -0.27
    ä¿ĥ
    -0.26
    itarian
    -0.25
     WRONG
    -0.24
    -Version
    -0.24
    astes
    -0.24
    otte
    -0.24
    为ä¸Ńå¿ĥ
    -0.24
    ç§ģä¸ĭ
    -0.24
    POSITIVE LOGITS
     their
    0.44
     your
    0.39
     the
    0.38
     all
    0.38
     a
    0.35
     its
    0.34
     any
    0.33
     half
    0.33
     these
    0.32
     one
    0.30
    Activations Density 0.894%

    No Known Activations