© 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. 31-TRANSCODER-HP
    7. 18804
    Prev
    Next
    INDEX
    Explanations

    code snippets

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ela
    -0.29
    -opacity
    -0.26
    L
    -0.26
     Brave
    -0.25
    OSP
    -0.25
    ystal
    -0.25
    åIJīåĪ©
    -0.24
     makeStyles
    -0.24
    itor
    -0.24
    äºŃ
    -0.24
    POSITIVE LOGITS
    ão
    0.30
    adero
    0.27
    handler
    0.27
    榴
    0.26
    ç»Ŀ
    0.25
     handling
    0.25
    Indented
    0.25
    ardi
    0.25
    ä¸ĢèĦļ
    0.24
    ifr
    0.24
    Activations Density 0.896%

    No Known Activations