© 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. 5153
    Prev
    Next
    INDEX
    Explanations

    treatments for conditions

    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
     avoided
    0.80
     Avoid
    0.79
    Avoid
    0.76
    避免
    0.72
    保证
    0.72
    避け
    0.72
    避
    0.71
    保
    0.71
     evaded
    0.71
     ensured
    0.67
    POSITIVE LOGITS
     addressing
    1.60
     address
    1.58
    Address
    1.54
     Address
    1.54
     addresses
    1.52
    address
    1.52
     Addresses
    1.43
    addresses
    1.43
    Addresses
    1.42
     Addressing
    1.40
    Activations Density 0.256%

    No Known Activations