© 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. 22-GEMMASCOPE-RES-131K
    4. 39733
    Prev
    Next
    INDEX
    Explanations

    color values like (0,0,0) or rgb

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

    The neuron never fires (all activations are zero), i.e. it’s a “dead” neuron that doesn’t detect any pattern.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    Œ
    -1.28
     ofte
    -1.23
    			
    -1.21
    الت
    -1.20
     hisz
    -1.20
    chait
    -1.17
    ſelf
    -1.17
    Character
    -1.15
     sied
    -1.15
    "
    -1.14
    POSITIVE LOGITS
     and
    1.65
    Motto
    1.36
     mehrere
    1.16
    Ilustra
    1.13
    Glit
    1.11
     their
    1.10
    Puss
    1.09
     edilen
    1.08
     zahlreiche
    1.07
     dieselben
    1.07
    Activations Density 0.006%

    No Known Activations