© 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. 34872
    Prev
    Next
    INDEX
    Explanations

    actions and emotions

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

    The neuron strongly activates on coordinating connectors—especially “and,” ampersands (“&”), and hyphens—i.e. tokens that link or join phrases.

    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
     intellectually
    -0.96
    腦
    -0.91
     pensamentos
    -0.90
     Mindset
    -0.88
    脑袋
    -0.88
    losse
    -0.88
     keenly
    -0.87
     mindset
    -0.85
     cognitive
    -0.82
    脑子
    -0.82
    POSITIVE LOGITS
     actions
    1.52
    行動
    1.38
     action
    1.27
     emotions
    1.21
     emotion
    1.14
     memory
    1.01
    actions
    0.93
     feeling
    0.93
     tindakan
    0.92
    Actions
    0.91
    Activations Density 0.051%

    No Known Activations