© 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. 22-GEMMASCOPE-RES-131K
    7. 39277
    Prev
    Next
    INDEX
    Explanations

    languages and dialects

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

    This neuron is acting as a positional encoder: it activates more strongly for tokens that occur further along in the text.

    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
    ૉ
    -1.35
     setup
    -1.35
    ]->
    -1.32
    –
    -1.26
    ]}$
    -1.23
    Kdo
    -1.23
    广场
    -1.23
     risks
    -1.20
    rootView
    -1.20
    联赛
    -1.18
    POSITIVE LOGITS
    revealed
    1.56
     ТАК
    1.41
     عاشقانه
    1.41
    футбо
    1.37
     religieuses
    1.34
    พูด
    1.32
    iress
    1.32
    暧
    1.30
    helped
    1.26
    
    1.26
    Activations Density 0.043%

    No Known Activations