© 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. 19057
    Prev
    Next
    INDEX
    Explanations

    say "like"

    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
    look
    -0.40
     look
    -0.39
    oble
    -0.30
     Look
    -0.30
     LOOK
    -0.29
     looks
    -0.29
    ILES
    -0.29
    глÑıд
    -0.28
     looked
    -0.28
    åĪĩ
    -0.28
    POSITIVE LOGITS
     much
    0.32
    much
    0.30
    ocator
    0.30
     mucho
    0.28
     Much
    0.28
    龸çİĭ
    0.25
    äºĨä¸įå°ij
    0.25
    ä¸īåĨľ
    0.25
    Much
    0.25
    å²Ķ
    0.24
    Activations Density 0.047%

    No Known Activations