© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 112436
    Prev
    Next
    INDEX
    Explanations

    want you to

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

    The neuron detects the infinitive marker “to” in direct-address commands or requests—i.e. the “you to <verb>” construction in phrases like “I want you to listen to…”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    ↵↵
    -1.80
    ?
    -1.68
    >
    -1.59
    </sup>
    -1.59
    ↵↵↵
    -1.59
    </h4>
    -1.58
     ]
    -1.58
     }
    -1.47
    it
    -1.41
    ]
    -1.41
    POSITIVE LOGITS
     jejich
    1.65
     belangrijk
    1.65
     první
    1.52
    💢
    1.50
     verschil
    1.50
    🆗
    1.48
    彼らの
    1.47
    〰
    1.45
    liegende
    1.45
     imma
    1.45
    Activations Density 0.033%

    No Known Activations