© 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. 90212
    Prev
    Next
    INDEX
    Explanations

    watching and relaxing

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

    The neuron strongly activates on passages of vivid, concrete, first-person experiential narration—that is, text describing personal actions, sensory details, and everyday objects.

    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
     slee
    -0.96
    fun
    -0.92
     >&
    -0.85
     partying
    -0.85
     bedtime
    -0.85
     skiing
    -0.85
     спать
    -0.82
    扱
    -0.79
    Fun
    -0.77
     snowboarding
    -0.77
    POSITIVE LOGITS
     watching
    2.09
     watch
    1.76
     observing
    1.66
     admiring
    1.58
     enjoying
    1.51
     Watching
    1.48
     listening
    1.48
    watching
    1.47
     watched
    1.41
    watch
    1.39
    Activations Density 0.024%

    No Known Activations