© 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. 95324
    Prev
    Next
    INDEX
    Explanations

    within

    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
    stone
    -0.28
    ister
    -0.27
    sumer
    -0.27
     '%$
    -0.26
    ivid
    -0.26
    Stone
    -0.24
    å¢ħ
    -0.24
    <Entry
    -0.23
    overe
    -0.23
     beside
    -0.23
    POSITIVE LOGITS
     reasonable
    0.38
    åIJĪçIJĨ
    0.31
     agreed
    0.29
    åįıè®®
    0.28
    æī¿è®¤
    0.28
     acceptable
    0.28
     harmony
    0.28
    åįıå®ļ
    0.28
    åIJĪçIJĨæĢ§
    0.28
     tack
    0.27
    Activations Density 0.887%

    No Known Activations