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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & 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. 51861
    Prev
    Next
    INDEX
    Explanations

    different variations and formats of the term "query" in request contexts

    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
     isIn
    -0.53
    enumii
    -0.53
    Marg
    -0.49
    toFloat
    -0.48
    Feb
    -0.44
     marginLeft
    -0.44
    SBATCH
    -0.44
    basicConfig
    -0.43
     removeFrom
    -0.43
    sheng
    -0.43
    POSITIVE LOGITS
     query
    1.49
     Query
    1.45
    query
    1.35
    Query
    1.27
     QUERY
    1.23
    QUERY
    1.12
     Queries
    1.08
     querying
    1.06
     queries
    1.05
    setQuery
    1.05
    Activations Density 0.012%

    No Known Activations