© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 37-GEMMASCOPE-2-TRANSCODER-262K
    7. 202985
    Prev
    Next
    INDEX
    Explanations

    The neuron detects patterns related to chemical structure descriptions, specifically hydrocarbons with single bonds and carbon chains. Phrases like "continuous carbon" and "straight-chain alkane" paired with numbers suggest it's identifying structural elements. The "Single Bonds Only - Saturated" text further reinforces this.Let's look for a unifying theme:* `MAX_ACTIVATING_TOKENS`: includes numbers (8, 0, 1, 2, 6) and descriptive words like "continuous", "only", "straight".* `TOKENS_AFTER_MAX_ACTIVATING_TOKEN`continuous carbonstraight chainsingle bonds

    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_37_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
     Agar
    0.46
    Agar
    0.45
     agar
    0.43
     agarose
    0.42
    Conv
    0.40
    莞
    0.40
     sandalwood
    0.40
     yg
    0.38
     geno
    0.38
     accidens
    0.38
    POSITIVE LOGITS
    复合
    0.35
    sold
    0.35
    ෂ්
    0.35
     hopper
    0.35
    Sold
    0.34
    сим
    0.34
     flea
    0.34
    ෝ
    0.34
    嘏
    0.34
     leased
    0.34
    Activations Density 0.001%

    No Known Activations