© 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. 32-TRANSCODER-HP
    7. 41763
    Prev
    Next
    INDEX
    Explanations

    Titles and names

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    3
    -0.30
    NS
    -0.29
    ä½ĵ
    -0.28
    ofire
    -0.27
    aces
    -0.27
    lets
    -0.27
    Sharp
    -0.26
    5
    -0.26
    stellar
    -0.25
    E
    -0.25
    POSITIVE LOGITS
     institutes
    0.26
    hic
    0.26
     reim
    0.25
    afil
    0.24
    ubes
    0.24
    æ¼¢
    0.24
    _hz
    0.23
     Coleman
    0.23
    ilen
    0.23
    ilege
    0.23
    Activations Density 0.000%

    No Known Activations