© 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. 18464
    Prev
    Next
    INDEX
    Explanations

    vitro, situ, or vivo

    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
     '
    -0.29
     '--
    -0.28
     '_
    -0.26
    _HEADERS
    -0.25
     '@
    -0.25
    -'
    -0.25
    åĩłåĪĨ
    -0.25
    è¿Ļç¬Ķ
    -0.25
    è¿ĻæĿ¡
    -0.25
     ';
    -0.25
    POSITIVE LOGITS
     setting
    0.29
    -setting
    0.28
     context
    0.28
     programming
    0.27
    设置
    0.27
     settings
    0.27
    backs
    0.26
     设置
    0.26
    	context
    0.26
    BOSE
    0.25
    Activations Density 0.069%

    No Known Activations