© 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. 32-TRANSCODER-HP
    7. 43813
    Prev
    Next
    INDEX
    Explanations

    ah

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ru
    -0.29
    éħ¥
    -0.27
     based
    -0.27
    alchemy
    -0.26
    based
    -0.26
     Lu
    -0.26
    ormal
    -0.26
    ruk
    -0.25
     involved
    -0.24
     Lois
    -0.23
    POSITIVE LOGITS
    ban
    0.28
     }};↵
    0.27
    ium
    0.27
    ç¾İçļĦ
    0.27
     PROVIDED
    0.27
    åīij
    0.27
    åĬį
    0.26
    èijĨ
    0.26
    kt
    0.26
    æĵĴ
    0.25
    Activations Density 0.004%

    No Known Activations