© 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. 31-GEMMASCOPE-RES-131K
    11. 63755
    Prev
    Next
    INDEX
    Explanations

    programming-related terms and constructs, particularly those associated with object management and state retrieval

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Armstrong
    -0.41
    igshid
    -0.39
     ig
    -0.38
    IG
    -0.38
     Mour
    -0.35
    катерина
    -0.35
     Verv
    -0.34
     محفوظة
    -0.34
     Ig
    -0.33
     dup
    -0.32
    POSITIVE LOGITS
     miniaturka
    0.71
     <<<<<<<<<<<<<<
    0.66
     estadounid
    0.61
     męska
    0.56
     étoient
    0.56
    ágenes
    0.54
     kapture
    0.54
     ainfi
    0.54
     trasparente
    0.53
    Erreferentziak
    0.52
    Activations Density 0.002%

    No Known Activations