© 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. 34-GEMMASCOPE-RES-131K
    4. 75678
    Prev
    Next
    INDEX
    Explanations

    AM prefix or abbreviation

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

    The neuron is consistently triggered by the uppercase letter sequence “AM” (e.g. in acronyms or chemical names).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     your
    -1.02
    djang
    -0.93
     Maternity
    -0.86
     都市
    -0.84
     童
    -0.84
     concreta
    -0.80
     attorneys
    -0.78
     exactly
    -0.78
     συμ
    -0.77
     simak
    -0.77
    POSITIVE LOGITS
     AM
    1.41
     آم
    1.18
    Am
    1.14
     Am
    1.05
     am
    1.02
     Ам
    0.94
    AM
    0.93
    AMP
    0.85
     ам
    0.83
     אמ
    0.81
    Activations Density 0.109%

    No Known Activations