© 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. 18691
    Prev
    Next
    INDEX
    Explanations

    Tightness

    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
    æľīåħ³
    -0.25
    é³Ħ
    -0.25
    åī©ä¸ĭ
    -0.25
    缸å½ĵ
    -0.25
    次è¦ģ
    -0.24
    èĽĻ
    -0.24
    cheduler
    -0.23
    涮
    -0.23
    caler
    -0.23
    olds
    -0.23
    POSITIVE LOGITS
    å¯Ĩ
    0.99
     closely
    0.84
    ç´§å¯Ĩ
    0.80
     tightly
    0.72
    å¯ĨåĪĩ
    0.70
     tight
    0.67
     intr
    0.66
    ç·Ĭ
    0.59
    ç´§
    0.56
    tight
    0.56
    Activations Density 24.149%

    No Known Activations