© 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. 21137
    Prev
    Next
    INDEX
    Explanations

    negations or suggestions not to engage in certain actions

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

    why not

    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
     gebou
    -0.36
     quæ
    -0.35
     opdracht
    -0.34
     procedimento
    -0.34
     boundaries
    -0.32
     procedimiento
    -0.32
     mijne
    -0.31
     procedure
    -0.31
     localização
    -0.30
     pertenecen
    -0.30
    POSITIVE LOGITS
    AnchorStyles
    0.79
     AssemblyProduct
    0.67
    rungsseite
    0.66
    Neden
    0.66
    WithIOException
    0.66
     AssemblyCulture
    0.65
    enumii
    0.65
    tagHelperRunner
    0.63
    Fancy
    0.62
    cashire
    0.60
    Activations Density 0.004%

    No Known Activations