© 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. 10-GEMMASCOPE-RES-131K
    7. 20392
    Prev
    Next
    INDEX
    Explanations

    try again

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

    The neuron activates on invitations to retry or repeat an action—most strongly on “try again” (and isolated “again”) tokens.

    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
     you
    -1.93
    ]
    -1.76
     }]
    -1.63
    蓁
    -1.55
     três
    -1.55
     zahlreichen
    -1.53
    érieurs
    -1.52
     выборе
    -1.51
    }$
    -1.50
    ጢ
    -1.50
    POSITIVE LOGITS
     with
    2.28
     similaire
    1.63
     through
    1.48
    我这
    1.45
    дента
    1.41
    𝒎
    1.40
     similar
    1.38
    的
    1.37
    Not
    1.36
     مشابه
    1.36
    Activations Density 0.008%

    No Known Activations