© 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. 22-GEMMASCOPE-RES-131K
    7. 26221
    Prev
    Next
    INDEX
    Explanations

    promises and claims

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

    The neuron fires on terms related to marketing and advertising (e.g. “advertises,” “marketing,” “hype,” “claims”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    lepiej
    -1.16
    tać
    -1.11
     recently
    -1.10
    providedIn
    -1.09
    bardziej
    -1.08
    AZIONI
    -1.06
    спасибо
    -1.06
     became
    -0.99
    RequestDto
    -0.99
    cić
    -0.98
    POSITIVE LOGITS
     promises
    1.59
     promise
    1.45
     promis
    1.37
     reinigen
    1.30
     selling
    1.30
     marketing
    1.29
     advertised
    1.28
     promised
    1.27
    Promise
    1.24
     claims
    1.21
    Activations Density 0.057%

    No Known Activations