© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-3-27B-IT
    3. 31-GEMMASCOPE-2-RES-262K
    4. 19829
    Prev
    Next
    INDEX
    Explanations

    problem-solving and error resolution

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

    phrases describing adverse events or problems—such as loss, harm, or failures—that imply the need for mitigation or intervention.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post/layer_31_width_262k_l0_medium
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    트에
    0.44
    ëren
    0.43
    asibility
    0.38
    erees
    0.38
    ཋ
    0.38
    クセ
    0.38
    트
    0.38
    LLO
    0.37
     considérons
    0.37
    人と
    0.37
    POSITIVE LOGITS
    解决方法
    0.63
     పరిష్
    0.58
     వెంటనే
    0.54
     çözüm
    0.52
    解决方案
    0.52
    慌
    0.51
     तुरंत
    0.50
     solution
    0.50
     해결
    0.50
     সমাধানের
    0.50
    Activations Density 0.137%

    No Known Activations