© 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. 18138
    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
    eted
    -0.27
    aders
    -0.26
    Mel
    -0.24
    Workers
    -0.24
    reck
    -0.23
     Mel
    -0.23
    cente
    -0.23
    TRS
    -0.23
    __/
    -0.23
     Shoe
    -0.23
    POSITIVE LOGITS
     morph
    0.27
    iceps
    0.27
    urf
    0.26
     toned
    0.26
     mor
    0.26
    æīĢ以æĪij们
    0.24
    ropri
    0.24
     skip
    0.24
     dee
    0.23
     trial
    0.23
    Activations Density 18.279%

    No Known Activations

    This feature has no known activations.