© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 24-GEMMASCOPE-TRANSCODER-16K
    7. 11147
    Prev
    Next
    INDEX
    Explanations

    this neuron seems to be catching a lot of common verbs and conjunctions, I can't tell what it is doing

    oai_token-act-pair · gemini-2.0-flash

    basing decisions on something

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_24/width_16k/average_l0_37
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.24.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Exacts
    -0.73
    WriteAttribute
    -0.71
    RTSC
    -0.70
    WebVitals
    -0.70
    ::_('
    -0.68
     Wikimédia
    -0.67
     للاسماء
    -0.66
    VIAF
    -0.66
     useAppContext
    -0.66
     ErrIntOverflow
    -0.65
    POSITIVE LOGITS
     right
    0.62
     far
    0.59
     back
    0.55
    zumaki
    0.55
     William
    0.54
     Bau
    0.53
     major
    0.52
     most
    0.52
     sister
    0.50
     around
    0.49
    Activations Density 8.322%

    No Known Activations