© 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. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 131k
    9. 31-GEMMASCOPE-RES-131K
    11. 13312
    Prev
    Next
    INDEX
    Explanations

    phrases related to social and economic inequalities

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

    political claims and accusations

    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
    ніципа
    -0.47
    Especially
    -0.43
     Especially
    -0.41
     kaynağından
    -0.37
     especialmente
    -0.36
     especially
    -0.36
     utford
    -0.35
    especially
    -0.35
     espe
    -0.34
     особенно
    -0.34
    POSITIVE LOGITS
     claiming
    0.85
     falsely
    0.81
     somehow
    0.79
     implying
    0.78
     яко
    0.75
     incorrectly
    0.74
     supposedly
    0.74
     claim
    0.73
     wrongly
    0.73
     blaming
    0.72
    Activations Density 0.125%

    No Known Activations