© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 40-GEMMASCOPE-2-RES-65K
    7. 48320
    Prev
    Next
    INDEX
    Explanations

    reward and incentives

    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_40_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
    堝
    0.40
    ката
    0.38
     declaratory
    0.38
     අධ
    0.37
     Ката
    0.37
    Prepar
    0.37
     invoking
    0.36
    zett
    0.36
    ഏ
    0.36
    უდ
    0.35
    POSITIVE LOGITS
     reward
    3.05
     rewards
    2.88
     Reward
    2.73
     rewarded
    2.70
    Reward
    2.67
    reward
    2.64
    奖励
    2.61
     Rewards
    2.45
     récomp
    2.45
    Rewards
    2.42
    Activations Density 0.117%

    No Known Activations