© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 75929
    Prev
    Next
    INDEX
    Explanations

    consider an example

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

    The neuron strongly activates on semantically “weighty” content words—i.e. uncommon nouns, verbs, or adjectives that carry high information content in the text.

    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
     gibi
    -1.28
    classed
    -1.22
     zwar
    -1.18
     queste
    -1.06
     dieser
    -1.05
     questo
    -1.03
    etheless
    -1.02
    řejmě
    -0.99
    ient
    -0.98
    vised
    -0.97
    POSITIVE LOGITS
     look
    4.03
     think
    3.63
     consider
    3.20
    Look
    3.03
     Look
    2.84
    Think
    2.72
     Think
    2.67
     Consider
    2.52
    look
    2.52
     remember
    2.52
    Activations Density 0.063%

    No Known Activations