© 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. 33177
    Prev
    Next
    INDEX
    Explanations

    numerical ratings or stars

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

    The neuron detects review‐style rating phrases—i.e. numerical or star scores (e.g. “gave the album a 9 out of 10,” “Rating: ★★★★,” “Our Score 8/10,” etc.).

    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
    sions
    -0.86
    thousands
    -0.82
    Aceptar
    -0.75
    𝒍
    -0.75
     확인
    -0.74
     ESL
    -0.72
    σμο
    -0.71
     Dobson
    -0.71
    分别是
    -0.71
    isand
    -0.70
    POSITIVE LOGITS
     rating
    2.94
     Rating
    2.50
    Rating
    2.48
     rated
    2.23
     ratings
    2.22
     RATING
    2.19
    rating
    2.16
     score
    2.03
    RATING
    1.95
     Ratings
    1.77
    Activations Density 0.037%

    No Known Activations