© 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. 65454
    Prev
    Next
    INDEX
    Explanations

    established firmly

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

    The neuron strongly activates on past‐tense or past‐participle verbs (e.g. “taken,” “failed,” “struggled,” “penetrated,” “ingrained”).

    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
    気味
    -0.82
     tertarik
    -0.77
    adden
    -0.76
    "..."
    -0.75
     sni
    -0.75
    eaus
    -0.74
    ⠟
    -0.73
     unification
    -0.73
    疇
    -0.71
     направ
    -0.70
    POSITIVE LOGITS
     root
    2.27
     roots
    2.25
     rooting
    2.08
     firmly
    1.90
    root
    1.89
     rooted
    1.88
    根
    1.87
     entrenched
    1.76
     established
    1.72
    roots
    1.70
    Activations Density 0.029%

    No Known Activations