© 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-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 127413
    Prev
    Next
    INDEX
    Explanations

    recombination, blushing, errors

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

    terms related to genetic recombination and molecular biology, especially words like "recombinase," "recombination," and "recombinant."

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron is keyed to the character sequence “recom,” i.e. the beginning of words like recombine, recombination, recombinant, etc.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    
    -2.89
    齎
    -2.73
    ↵↵
    -2.61
    恫
    -2.56
    錚
    -2.48
    逦
    -2.42
    鹚
    -2.42
    忖
    -2.36
    顰
    -2.33
    中毒
    -2.25
    POSITIVE LOGITS
    0
    4.56
    4
    4.16
    3
    3.91
    1
    3.75
    6
    3.73
    8
    3.58
    9
    3.53
    7
    3.39
    2
    3.31
    Why
    3.11
    Activations Density 0.003%

    No Known Activations