© 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. 34-GEMMASCOPE-RES-131K
    7. 78735
    Prev
    Next
    INDEX
    Explanations

    intentions and attitudes

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

    This neuron selectively activates on content words in the main body of academic prose (e.g. scientific‐abstract sentences) rather than on section headings, metadata, or code.

    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
     Carney
    -0.78
    境界
    -0.77
     darb
    -0.73
     eing
    -0.73
    SignOut
    -0.73
    몽
    -0.69
     recibir
    -0.69
    emia
    -0.69
     poverty
    -0.68
     Advertiser
    -0.68
    POSITIVE LOGITS
     intention
    1.31
     Intention
    1.17
     continuance
    1.16
     acceptance
    1.10
     intentions
    1.05
     attitude
    0.95
     antecedents
    0.95
     adoption
    0.93
    Acceptance
    0.93
    intention
    0.91
    Activations Density 0.025%

    No Known Activations