© 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. 43039
    Prev
    Next
    INDEX
    Explanations

    Code, education, and gaming scores

    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
     Regel
    -0.25
    enos
    -0.24
    érica
    -0.24
    .Inject
    -0.24
    IÃĵN
    -0.23
    èĥľ
    -0.23
    @Inject
    -0.23
    æŀŀ
    -0.23
    enÃŃ
    -0.23
     dereg
    -0.23
    POSITIVE LOGITS
     tested
    0.34
    ħĮ
    0.29
     scientific
    0.29
    åĬłçĽŁ
    0.29
     drill
    0.28
     AGE
    0.28
    sc
    0.27
    .grade
    0.27
     tests
    0.27
     grade
    0.27
    Activations Density 0.001%

    No Known Activations