Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 80353
    Prev
    Next
    INDEX
    Explanations

    specific biochemical or genetic markers

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

    short chemical prefixes

    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-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
    ılıyor
    -0.52
    i
    -0.48
    ığını
    -0.46
    ి
    -0.45
    しまって
    -0.44
    tığı
    -0.43
    ıyordu
    -0.43
    joni
    -0.42
    bourne
    -0.42
    imdi
    -0.42
    POSITIVE LOGITS
    ا
    0.73
    ak
    0.72
    ad
    0.72
    ab
    0.71
    ac
    0.71
    ap
    0.70
    ag
    0.68
    acal
    0.68
    av
    0.67
    AC
    0.65
    Activations Density 0.168%

    No Known Activations