© 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. 18856
    Prev
    Next
    INDEX
    Explanations

    a

    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
    åį«
    -0.28
    leared
    -0.27
    çıŃ
    -0.27
    æįİ
    -0.26
     genu
    -0.26
    ìĥģ
    -0.26
    zzo
    -0.26
    bum
    -0.25
    ine
    -0.25
    prises
    -0.25
    POSITIVE LOGITS
    jar
    0.29
    å¼¹
    0.25
    ghan
    0.25
    被æīĵ
    0.25
    éķĩ
    0.24
    rax
    0.24
    fsp
    0.24
    aco
    0.23
     IS
    0.23
    KS
    0.23
    Activations Density 5.487%

    No Known Activations