© 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
    3. Gemma-3-27B-IT
    5. 41-GEMMASCOPE-2-RES-262K
    7. 32907
    Prev
    Next
    INDEX
    Explanations

    escalation and de-escalation

    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_all/layer_41_width_262k_l0_big
    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
     Polyp
    0.40
    addle
    0.39
     මත
    0.38
    damageCount
    0.35
    דות
    0.35
     photochemical
    0.35
     Polyg
    0.35
    ꯠ
    0.35
     purg
    0.34
     Formulas
    0.34
    POSITIVE LOGITS
     escalation
    1.70
    escal
    1.52
     escalate
    1.52
     escal
    1.48
     Escal
    1.47
     de
    1.45
     escalated
    1.45
     escalating
    1.41
     De
    1.19
     escalar
    1.09
    Activations Density 0.002%

    No Known Activations