© 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. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 131k
    9. 31-GEMMASCOPE-RES-131K
    11. 11901
    Prev
    Next
    INDEX
    Explanations

    phrases related to customization and personal expression

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    perfect, beautiful, stunning

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    principalTable
    -0.71
     ProtoMessage
    -0.69
    AddTagHelper
    -0.66
    queryInterface
    -0.66
     ivelany
    -0.61
    enderror
    -0.60
    (!__
    -0.56
    RegressionTest
    -0.55
     Roskov
    -0.55
    estacks
    -0.53
    POSITIVE LOGITS
     beautiful
    0.63
     perfect
    0.63
     stunning
    0.57
     masterpiece
    0.52
     satisfactory
    0.52
    beautiful
    0.51
     amazing
    0.51
    完美
    0.50
     perfecta
    0.49
     magnifique
    0.49
    Activations Density 0.020%

    No Known Activations