© 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. 116713
    Prev
    Next
    INDEX
    Explanations

    code blocks ending

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

    The neuron fires strongly on unusually large whitespace runs or big indentation/blank‐line blocks.

    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
     though
    -0.90
     like
    -0.87
     similar
    -0.85
    Недостатки
    -0.83
     suitable
    -0.83
     wherever
    -0.81
    Ofic
    -0.79
     refugiados
    -0.79
    饧
    -0.79
     caoutchouc
    -0.78
    POSITIVE LOGITS
        
    1.28
            
    0.91
     }
    0.89
    },
    0.88
    }
    0.88
          
    0.85
    到底
    0.84
    };
    0.82
     Else
    0.82
    quely
    0.81
    Activations Density 0.004%

    No Known Activations