© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 109742
    Prev
    Next
    INDEX
    Explanations

    "I think I" thinking

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    believe
    -1.16
     berpikir
    -1.16
     pensare
    -1.11
     think
    -1.09
    дума
    -1.05
     думаю
    -1.05
    认为
    -1.05
     tror
    -1.03
     pensar
    -1.02
    Tipps
    -1.01
    POSITIVE LOGITS
     mentioned
    1.23
     have
    1.13
    see
    1.04
     heard
    1.03
     should
    1.02
     speak
    1.01
     need
    1.00
     will
    0.94
     said
    0.93
     seen
    0.92
    Activations Density 0.022%

    No Known Activations