© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 38051
    Prev
    Next
    INDEX
    Explanations

    package metadata and abstracts

    np_acts-logits-general · gemini-2.5-flash-lite

    key-value pairs in package configuration files like package.json, gemspec files, and setup configurations.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron fires strongly on the field‐name tokens used in software package manifests (e.g. “name,” “version,” “description,” “license,” “repository,” etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     föl
    -0.75
     nhung
    -0.73
    leisure
    -0.71
    BaseAdapter
    -0.71
    quième
    -0.70
     espar
    -0.68
     Whale
    -0.68
    forgotten
    -0.67
     الرحمن
    -0.66
     standaard
    -0.66
    POSITIVE LOGITS
     Mollus
    0.73
    તા
    0.69
     đấu
    0.67
    目が
    0.65
     Cakes
    0.65
    шная
    0.65
    HttpClientModule
    0.64
    antin
    0.64
     अर्थ
    0.64
     Campan
    0.63
    Activations Density 0.049%

    No Known Activations