© 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. 38485
    Prev
    Next
    INDEX
    Explanations

    Police investigate, Protein, degradation of

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

    code-related syntax elements in technical documentation or source code files.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron responds to rare or infrequent tokens—things like uncommon punctuation (e.g. “{”), standalone numbers, or unusual letter sequences.

    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
    ↵
    -1.35
     to
    -0.91
    !
    -0.86
    -0.86
    ;

    -0.84
    цей
    -0.80
    ]),

    -0.78
    getIntent
    -0.76
    ))}
    -0.74
    ？（
    -0.74
    POSITIVE LOGITS
    讥
    0.99
     supplé
    0.91
    ziehung
    0.85
    awtextra
    0.85
     gna
    0.84
    ibrator
    0.83
    freunde
    0.82
    队的
    0.81
    })+
    0.81
    isierte
    0.81
    Activations Density 0.000%
    No Known Activations