© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-3-27B-IT
    3. 27-GEMMASCOPE-2-TRANSCODER-262K
    4. 196568
    Prev
    Next
    INDEX
    Explanations

    Pig Latin examples

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

    # Explanation of neuron 4 behavior: the main thing this neuron does is find **numeric values and date/time-related tokens** that appear in structured documents or formatted lists.

    oai_token-act-pair · claude-4-5-haikuTriggered by @enricobottazzi

    dates formatted as MM/DD/YYYY.

    oai_token-act-pair · deepseek-v3Triggered by @enricobottazzi
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/transcoder_all/layer_27_width_262k_l0_small_affine
    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
     Psychic
    0.41
     Daily
    0.38
    5
    0.37
    Query
    0.37
     Цена
    0.36
     Price
    0.36
     Page
    0.36
    śred
    0.36
     Companies
    0.36
     Университе
    0.36
    POSITIVE LOGITS
    燁
    0.36
    եր
    0.35
    寒
    0.35
     !_
    0.35
     nonlinearity
    0.33
    BleStatus
    0.32
     végétation
    0.32
    供
    0.32
     calor
    0.32
    яти
    0.32
    Activations Density 0.000%

    No Known Activations