© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 84612
    Prev
    Next
    INDEX
    Explanations

    job termination

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

    The main thing this neuron does is detect words and phrases describing someone being fired, relieved, or otherwise terminated from a position.

    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
    刻意
    -0.91
     прио
    -0.83
     disallowed
    -0.79
     serotonin
    -0.79
    Withdrawal
    -0.78
     withdrawal
    -0.77
    uhnya
    -0.77
    UNKNOWN
    -0.76
     запре
    -0.75
    сели
    -0.75
    POSITIVE LOGITS
     dismissal
    1.84
     fired
    1.50
     termination
    1.48
    Dismiss
    1.36
     firing
    1.32
    dismiss
    1.30
     Fired
    1.27
     Dismiss
    1.24
     Termination
    1.24
     uncer
    1.22
    Activations Density 0.033%

    No Known Activations