© 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. 82944
    Prev
    Next
    INDEX
    Explanations

    dating back

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

    The neuron responds most strongly to very common high‐frequency function words (e.g. “the,” “to,” pronouns, conjunctions, 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
     außerdem
    -1.78
    しつつ
    -1.34
     doubtless
    -1.27
    事实上
    -1.25
     Außerdem
    -1.25
     żeby
    -1.16
     offenbar
    -1.13
     einerseits
    -1.13
     thence
    -1.13
    σιο
    -1.11
    POSITIVE LOGITS
     predomin
    1.18
     somit
    1.12
    FYI
    1.11
     monies
    1.08
     lainnya
    1.05
     purposely
    1.01
    此
    0.97
    ekk
    0.93
    ]^{
    0.93
    نوع
    0.93
    Activations Density 0.160%

    No Known Activations