© 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. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 31-GEMMASCOPE-RES-131K
    6. 12394
    Prev
    Next
    INDEX
    Explanations

    comparative expressions and metrics across various contexts

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

    entirety and combined

    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
     disambiguazione
    -0.56
    Personendaten
    -0.49
     particularly
    -0.45
     Different
    -0.45
     especially
    -0.44
    InjectAttribute
    -0.44
     особенно
    -0.43
    듯
    -0.42
     Especially
    -0.42
     khususnya
    -0.42
    POSITIVE LOGITS
     entire
    0.98
     ENTIRE
    0.85
     combined
    0.84
     COMBIN
    0.84
    entire
    0.80
    combined
    0.77
     Entire
    0.74
     gesamten
    0.73
     persino
    0.71
     addirittura
    0.69
    Activations Density 0.048%

    No Known Activations