© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 118971
    Prev
    Next
    INDEX
    Explanations

    understand or misunderstanding

    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
     Coen
    -0.76
    翻
    -0.73
    entru
    -0.72
     позво
    -0.72
     pensando
    -0.71
    waitFor
    -0.71
    厉
    -0.69
    makers
    -0.68
     ジェル
    -0.68
     reun
    -0.68
    POSITIVE LOGITS
    stand
    2.52
     understanding
    2.47
    understanding
    2.30
     stand
    2.25
     understand
    2.19
    standing
    2.14
     Understanding
    2.11
    Understanding
    2.11
    stands
    2.09
    stood
    2.08
    Activations Density 0.014%

    No Known Activations