© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 65867
    Prev
    Next
    INDEX
    Explanations

    adventures and trouble

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

    The neuron detects words that signal plot‐driving actions or narrative events—verbs and key nouns that describe a character’s journey or what happens in the story.

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

    descriptions of protagonists engaging in adventures or journeys and getting into or out of trouble within narrative summaries.

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

    narrative text describing a character's journey, quest, or series of adventures.

    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
     jego
    -1.12
     whereas
    -1.05
     plateado
    -1.03
    ołnier
    -1.00
     weißem
    -0.99
    tômes
    -0.99
     terwijl
    -0.97
     granic
    -0.96
    ficulty
    -0.95
    fifty
    -0.95
    POSITIVE LOGITS
    Setelah
    1.22
    ится
    1.17
    通过
    1.16
    非常
    1.11
     through
    1.09
     tentar
    1.09
    بعد
    1.09
    Szer
    1.07
    macam
    1.06
    karena
    1.06
    Activations Density 0.067%

    No Known Activations