© 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-3-4B-IT
    5. 29-GEMMASCOPE-2-RES-65K
    7. 5309
    Prev
    Next
    INDEX
    Explanations

    key improvements error handling

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-4b-it/resid_post/layer_29_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
     можете
    0.88
    かもしれません
    0.83
     يمكنك
    0.75
    :)
    0.75
     możemy
    0.74
     might
    0.73
    它可以
    0.73
    :')
    0.72
     могла
    0.72
    might
    0.71
    POSITIVE LOGITS
     Included
    1.18
     included
    1.08
    Included
    1.07
     improves
    1.02
     improved
    1.02
    included
    0.98
    Improved
    0.97
     Improves
    0.97
     corrected
    0.97
    Added
    0.97
    Activations Density 0.095%

    No Known Activations