© 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. 30-TRANSCODER-HP
    7. 95306
    Prev
    Next
    INDEX
    Explanations

    Nerves and arteries

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     democr
    -0.28
     ub
    -0.27
    xc
    -0.26
    æĻļ
    -0.26
     xc
    -0.24
    .dict
    -0.24
     democratic
    -0.24
    åį«
    -0.24
    Echo
    -0.24
    ubi
    -0.23
    POSITIVE LOGITS
    IAM
    0.30
    梦è§ģ
    0.28
    AIM
    0.28
    iedo
    0.26
    bett
    0.26
     bé
    0.26
    lein
    0.25
    oped
    0.24
    oze
    0.23
     Aad
    0.23
    Activations Density 0.107%

    No Known Activations