© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 122872
    Prev
    Next
    INDEX
    Explanations

    specific morphological or syntactic patterns in text, particularly those concerning grammatical constructs and relationships

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

    installing verifying replicating

    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
     betweenstory
    -1.01
    zzleHttp
    -0.80
    Portály
    -0.77
     nonUne
    -0.70
     Felsen
    -0.70
     AssemblyCulture
    -0.70
    AISSEE
    -0.70
    ItemBackground
    -0.67
    󠁴
    -0.66
    脚注の使い方
    -0.65
    POSITIVE LOGITS
     प
    0.45
     
    0.42
     fire
    0.42
     Ep
    0.42
     gr
    0.41
     ep
    0.41
     الط
    0.40
     ش
    0.40
     Comp
    0.40
     hal
    0.39
    Activations Density 0.014%

    No Known Activations