© 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. 54590
    Prev
    Next
    INDEX
    Explanations

    references to images and artwork in various contexts

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

    names, versions, amounts

    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
    GEBURTSDATUM
    -0.74
    SharedDtor
    -0.50
     tartalomajánló
    -0.48
    webElementXpaths
    -0.48
    出版年
    -0.48
    iconque
    -0.47
    rungsseite
    -0.47
    adpleegd
    -0.46
    帖最后由
    -0.45
     esprits
    -0.45
    POSITIVE LOGITS
     themselves
    0.46
     teils
    0.43
     these
    0.41
    themselves
    0.40
     jeweils
    0.40
    一个个
    0.38
    UnitTesting
    0.38
     jeweiligen
    0.37
     descriptions
    0.37
     each
    0.36
    Activations Density 0.108%

    No Known Activations