© 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. 19358
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
     addressing
    -0.28
     addressed
    -0.25
    core
    -0.25
    acted
    -0.25
    æĢ»ä½ĵè§ĦåĪĴ
    -0.25
    tf
    -0.25
    erno
    -0.25
    quired
    -0.25
    ignet
    -0.25
     exposures
    -0.24
    POSITIVE LOGITS
    legate
    0.30
    .blogspot
    0.26
    leta
    0.26
    æĪŁ
    0.26
    .twitter
    0.25
    å¹¿åľº
    0.25
    Jose
    0.24
    æľī大éĩıçļĦ
    0.24
    æµ·éĩı
    0.24
    LETE
    0.24
    Activations Density 0.124%

    No Known Activations

    This feature has no known activations.