© 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. 18187
    Prev
    Next
    INDEX
    Explanations

    non-English language

    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
    ibus
    -0.33
    itations
    -0.31
    iw
    -0.31
    ifica
    -0.28
    idade
    -0.28
    iston
    -0.28
    igi
    -0.28
    инг
    -0.27
    ific
    -0.27
    çªŁ
    -0.27
    POSITIVE LOGITS
    erte
    0.35
    ção
    0.33
    erten
    0.33
    ert
    0.32
    erts
    0.30
    è³
    0.30
    ng
    0.30
    ços
    0.28
    ções
    0.28
    NG
    0.26
    Activations Density 16.643%

    No Known Activations