© 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 - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 13762
    Prev
    Next
    INDEX
    Explanations

    statements and quotes from individuals, particularly focusing on reported speech

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

    Attribution, often with "said"

    np_acts-logits-general · gemini-2.0-flash

    attribution of statements

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     betweenstory
    -0.85
     ब्रेकडाउन
    -0.74
     resourceCulture
    -0.74
    WebServlet
    -0.72
     ویکی‌پدی
    -0.71
    IUrlHelper
    -0.69
    uxxxx
    -0.68
     []*
    -0.66
     незавершена
    -0.66
    ReusableCell
    -0.66
    POSITIVE LOGITS
     pihaknya
    0.42
     bahwa
    0.41
     gdyby
    0.39
     bahawa
    0.39
     said
    0.38
     kwamba
    0.37
    Menurutnya
    0.36
     that
    0.35
     wpły
    0.34
     fehl
    0.34
    Activations Density 0.148%

    No Known Activations