© 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 - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 9243
    Prev
    Next
    INDEX
    Explanations

    repetitive structures or patterns in language

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

    "Each" or "every" followed by a noun

    np_acts-logits-general · gemini-2.0-flash

    each ... is

    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
    pkey
    -0.41
     general
    -0.36
     hiç
    -0.36
    Unset
    -0.36
     bazı
    -0.35
     inconvenient
    -0.35
     disposed
    -0.35
     unwilling
    -0.35
     most
    -0.34
     many
    -0.34
    POSITIVE LOGITS
    Each
    0.80
     Each
    0.79
    Every
    0.77
     Every
    0.76
    Chaque
    0.74
     each
    0.74
     Jedes
    0.73
    每一次
    0.71
     every
    0.71
    every
    0.70
    Activations Density 0.384%

    No Known Activations