© 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
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 131k
    9. 20-GEMMASCOPE-RES-131K
    11. 76343
    Prev
    Next
    INDEX
    Explanations

    mentions of models and document-related concepts in programming contexts

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

    state and change operations

    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
    новниш
    -0.60
    ActionButton
    -0.58
    Източници
    -0.56
    tvguidetime
    -0.54
    cheduler
    -0.50
     EINVAL
    -0.50
     Stalker
    -0.50
    +#+#
    -0.49
    IsInitialized
    -0.49
    eleste
    -0.48
    POSITIVE LOGITS
    TextWatcher
    0.49
    errHandler
    0.36
     regla
    0.36
    теристика
    0.34
     peintre
    0.34
    Personendaten
    0.34
    териалы
    0.33
     Wikimedijinoj
    0.32
     inför
    0.32
     modific
    0.32
    Activations Density 0.001%

    No Known Activations