© 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. 18-TRANSCODER-HP
    7. 89191
    Prev
    Next
    INDEX
    Explanations

    the

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_18.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.18.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    inee
    -0.28
     soon
    -0.28
     frequently
    -0.26
    angs
    -0.25
    eward
    -0.25
    .cum
    -0.25
    quee
    -0.25
    éķ¿æŶéĹ´
    -0.24
    pons
    -0.24
     Naked
    -0.24
    POSITIVE LOGITS
    éļĶçĿĢ
    0.30
     Trib
    0.29
    ä¸ºä½ł
    0.27
    å½ĵæĪij
    0.27
    å¯¹ä½ł
    0.25
    æ±
    0.25
    ä¿¡èµĸ
    0.25
    临
    0.25
    对æĪij
    0.25
     trib
    0.24
    Activations Density 4.614%

    No Known Activations