© 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. 31-GEMMASCOPE-2-RES-262K
    7. 6041
    Prev
    Next
    INDEX
    Explanations

    solely

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

    the neuron activates on tokens that are part of the assistant/model's responses (i.e., text produced by the model rather than user input).

    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_31_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
    (
    0.23
    า
    0.20
    on
    0.19
    を使用
    0.18
    wiches
    0.18
    <h2>
    0.18
     (
    0.18
    et
    0.18
    }(
    0.18
    4
    0.17
    POSITIVE LOGITS
     Frankly
    0.17
     जानता
    0.17
     nemmeno
    0.17
    ócio
    0.17
     Vocational
    0.16
     circunstancias
    0.16
     Knowing
    0.16
     દે
    0.16
     urgently
    0.15
     Hui
    0.15
    Activations Density 3.851%

    No Known Activations