© 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. 95854
    Prev
    Next
    INDEX
    Explanations

    designs, shapes, styles

    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
     shape
    -1.28
    å½¢çĬ¶
    -1.21
    å½¢
    -1.17
    å½¢æĢģ
    -1.15
    shape
    -1.13
     shapes
    -1.13
     Shape
    -1.10
    Shape
    -1.01
    _shape
    -0.98
     shaped
    -0.96
    POSITIVE LOGITS
    ì¹ĺ
    0.28
    eva
    0.24
    æĺĮ
    0.24
    å¥ĩ迹
    0.24
    asted
    0.24
    é¼ĵ
    0.24
    AGES
    0.23
    prises
    0.23
    ADR
    0.23
    客æľį
    0.23
    Activations Density 0.906%

    No Known Activations