© 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. 24105
    Prev
    Next
    INDEX
    Explanations

    instances of replacing one entity with another

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

    replacing with

    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
     beyond
    -0.52
    tangentMode
    -0.51
    uidado
    -0.48
     both
    -0.45
     Further
    -0.45
     cả
    -0.44
     Both
    -0.42
    getragen
    -0.42
    tagHelper
    -0.42
    both
    -0.41
    POSITIVE LOGITS
     replacements
    0.78
     replaced
    0.77
     equivalents
    0.76
     equivalent
    0.76
     substitutes
    0.74
     something
    0.73
     alternatives
    0.72
     ones
    0.71
    equivalent
    0.68
     replacement
    0.68
    Activations Density 0.076%

    No Known Activations