© 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
    2. Gemma-3-27B-IT
    3. 27-GEMMASCOPE-2-TRANSCODER-262K
    4. 198602
    Prev
    Next
    INDEX
    Explanations

    The user asks for a concise explanation (3-20 words) of a neuron's behavior, focusing on patterns in lists.I need to analyze:1. **MAX_ACTIVATING_TOKENS**: 'etc', 'etc', 'etc', 'c', 'etc', 'etc', 'on', 'forth', 'etc', 'etc'2. **TOKENS_AFTER_MAX_ACTIVATING_TOKEN**: '.', '.\'', 'also', '"', '."', ',', '":', '.'3. **TOP_POSITIVE_LOGITS**: 'aggress', '행사', 'rosa', 'LCA', '%.The provided lists show a strong presence of "etc" in MAX_ACTIVATING_TOKENS, and "on", "forth" which often complete "et cetera". The TOKENS_AFTER_MAX_ACTIVATING_TOKEN are typical punctuation or small words that follow such phrases. Looking at TOP_ACTIVATING_TEXTS, "etc." appears multiple times in contexts like "and so on, or 'etc.'" and "phrases like 'and', 'or', 'etc.'". This suggests the neuron is related to the concept of listing continuation or examples.The TOP_POSITIVE_LOGITS are diverse and don't immediately form a clear pattern related to "etc.". However, the instruction says "not ALL top activating texts/tokens have to match the exact same pattern, but a majority should."Given the strong signal from MAX_ACTIVATING_TOKENS and TOP_ACTIVATING_TEXTS, the neuron is clearly associated with the use of "etc." for listing continuation or implying more items.I need to express this concisely.listing continuation

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    变
    0.43
    ချက်
    0.42
    我又
    0.41
    ㇁
    0.40
    ালী
    0.40
    선을
    0.40
    沸
    0.39
    ไม่ใช่
    0.38
     השי
    0.38
    我和
    0.38
    POSITIVE LOGITS
     aggress
    0.43
     행사
    0.39
     rosa
    0.38
     LCA
    0.38
     %.
    0.38
     pela
    0.37
     accumulate
    0.36
     flake
    0.36
     that
    0.36
     cropped
    0.36
    Activations Density 0.000%

    No Known Activations