© 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
    5. 31-GEMMASCOPE-2-RES-262K
    7. 218549
    Prev
    Next
    INDEX
    Explanations

    until browned

    np_acts-logits-general · gemini-2.5-flash-lite

    capturing context clues

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-pt/resid_post/layer_31_width_262k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    為了
    0.51
    ਾਸ
    0.51
    美国
    0.50
    José
    0.50
    東京
    0.49
    Türk
    0.49
    會
    0.49
    伨
    0.49
    虢
    0.47
    为了
    0.47
    POSITIVE LOGITS
     Affidavit
    0.46
     chatbots
    0.45
     hugely
    0.44
     insists
    0.44
     Progressives
    0.44
     said
    0.44
     advises
    0.44
     exacerbated
    0.43
     briefed
    0.43
     confided
    0.43
    Activations Density 0.001%

    No Known Activations