© 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. 20-GEMMASCOPE-RES-131K
    6. 12185
    Prev
    Next
    INDEX
    Explanations

    names of major technology companies, particularly Amazon and Apple

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

    company names followed by related terms

    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
     يتيمه
    -0.63
    SharedCtor
    -0.51
     disambiguazione
    -0.46
    webElementXpaths
    -0.42
    niająca
    -0.38
     betweenstory
    -0.38
    UnusedPrivate
    -0.38
    LookAnd
    -0.37
     invokingState
    -0.37
    eraard
    -0.35
    POSITIVE LOGITS
    󠁣
    0.63
     houſe
    0.63
    ſelf
    0.60
     ſtate
    0.57
     plagio
    0.57
     alpina
    0.56
     myſelf
    0.55
     BoxDecoration
    0.55
     pleaſure
    0.55
    Чыгана
    0.53
    Activations Density 0.035%

    No Known Activations