© 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-3-27B-IT
    5. 40-GEMMASCOPE-2-RES-262K
    7. 18006
    Prev
    Next
    INDEX
    Explanations

    contrastive conjunctions

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

    A strong signal for tokens that are part of the model/assistant's generated responses (often formatted/structured text like headings, emphasis, and explanatory sentences).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post/layer_40_width_262k_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
    трите
    0.40
    йте
    0.38
    ਤੇ
    0.38
    變化
    0.38
    ለ
    0.38
     রাখেন
    0.37
     використо
    0.37
    rowth
    0.37
     রাখে
    0.36
    conversation
    0.36
    POSITIVE LOGITS
     getting
    0.86
     putting
    0.81
     sending
    0.75
     placing
    0.75
     owning
    0.74
     adding
    0.73
     importing
    0.73
     deploying
    0.71
     having
    0.71
     renting
    0.71
    Activations Density 0.064%

    No Known Activations