© 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-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 91433
    Prev
    Next
    INDEX
    Explanations

    hard, firm, stiff

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

    The neuron strongly activates on common function words (articles, conjunctions, pronouns, auxiliary verbs) and boilerplate terms typical of technical/patent-style descriptions.

    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
    🕌
    -0.85
     снег
    -0.78
    Läh
    -0.78
     processos
    -0.76
    товой
    -0.73
    edited
    -0.73
    fret
    -0.73
    ╭
    -0.72
     Straus
    -0.71
    Exped
    -0.71
    POSITIVE LOGITS
    硬
    3.91
     harden
    3.67
     hard
    3.67
     hardening
    3.53
     hardened
    3.52
     hardness
    3.50
    hard
    3.30
     firm
    3.25
     stiff
    3.13
     harder
    3.08
    Activations Density 0.096%

    No Known Activations