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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    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. 101384
    Prev
    Next
    INDEX
    Explanations

    instances of contrast or opposition in arguments or statements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    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
    Билгалдахарш
    -0.49
     ainfi
    -0.45
    HtmlAttribute
    -0.42
     prisonniers
    -0.42
     nggak
    -0.41
    ientras
    -0.41
    čního
    -0.41
    ❋
    -0.41
     étoient
    -0.40
     chré
    -0.39
    POSITIVE LOGITS
     including
    0.58
    endregion
    0.57
    copyWith
    0.52
     Including
    0.50
     včetně
    0.50
     waaronder
    0.50
    Including
    0.50
    それだけ
    0.49
     include
    0.49
    包括
    0.48
    Activations Density 0.008%

    No Known Activations