© 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. 34-GEMMASCOPE-RES-131K
    4. 33184
    Prev
    Next
    INDEX
    Explanations

    Augmented reality

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

    The neuron fires strongly on words that occur in idiomatic multiword expressions of separation or disconnection (e.g. “far off from reality,” “utterly divorced from…,” “out of sync with…”).

    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
    শে
    -0.80
    lň
    -0.77
     suất
    -0.77
    enské
    -0.77
    useEffect
    -0.76
    ("")){
    -0.76
    </tr>
    -0.76
    ジュ
    -0.73
     ObjectParameter
    -0.73
     });
    
    -0.73
    POSITIVE LOGITS
     reality
    8.63
    reality
    6.84
    Reality
    6.09
     Reality
    5.84
     realidad
    5.69
     realities
    5.66
     réalité
    5.31
     realidade
    4.88
     realtà
    4.53
     actuality
    4.03
    Activations Density 0.085%

    No Known Activations