© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 73370
    Prev
    Next
    INDEX
    Explanations

    stasis, toxins, hoppers, stroboscopic

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

    The neuron fires on the initial few letters of longer, multi‐syllable technical or domain-specific terms (e.g. “di-” in dihydroergotamine, “st-” in stasis, “ar-” in arthrofibrosis).

    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
    ですが
    -1.10
     которому
    -1.10
    と
    -1.08
     feind
    -1.07
    ̀i
    -1.05
    하며
    -1.02
    ly
    -0.98
    もある
    -0.97
    olk
    -0.95
    などは
    -0.95
    POSITIVE LOGITS
     its
    1.35
    corrhi
    1.21
    に追加
    1.21
    様な
    1.19
    の準備
    1.18
     reservados
    1.18
    ようになりました
    1.14
    ナメント
    1.14
     emocion
    1.12
    のご紹介
    1.12
    Activations Density 0.374%

    No Known Activations