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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 31-GEMMASCOPE-2-RES-65K
    7. 11715
    Prev
    Next
    INDEX
    Explanations

    beforehand

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post/layer_31_width_65k_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
    ERRORTYPE
    0.41
     deont
    0.39
     PA
    0.38
     علت
    0.38
     successivamente
    0.37
     uygulam
    0.36
     yi
    0.35
     α
    0.35
     लड़
    0.35
    oxicity
    0.35
    POSITIVE LOGITS
     미리
    1.70
     заранее
    1.64
     beforehand
    1.56
    提前
    1.46
    事前に
    1.39
    事先
    1.35
    事前
    1.33
    准备
    1.23
     prepares
    1.21
     사전
    1.21
    Activations Density 0.322%

    No Known Activations