© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 7459
    Prev
    Next
    INDEX
    Explanations

    expressing strong emotions

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

    This neuron detects words related to venting or ranting—instances of emotional outpouring.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    language indicating emotional venting, rants, or outbursts and the release of frustration.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408

    words related to expressing strong emotions or releasing pent-up feelings (venting, letting off steam, outbursts, rants).

    oai_token-act-pair · claude-4-5-sonnetTriggered 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
     begle
    -1.69
     Now
    -1.60
     recognised
    -1.52
     augen
    -1.50
     reason
    -1.50
     beobach
    -1.45
     deliver
    -1.42
     At
    -1.42
    which
    -1.42
    IS
    -1.40
    POSITIVE LOGITS
     něj
    1.80
    鸶
    1.77
    珧
    1.58
     possi
    1.57
     konkret
    1.52
     стоковые
    1.46
     passende
    1.42
    Suka
    1.41
    suchen
    1.39
     různých
    1.38
    Activations Density 0.010%

    No Known Activations