© 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. 31-GEMMASCOPE-RES-131K
    6. 10260
    Prev
    Next
    INDEX
    Explanations

    proper names of individuals and brands related to various contexts

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

    personal names and associated words

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    sc
    -0.38
    quite
    -0.36
    so
    -0.34
    cl
    -0.33
    (!)
    -0.32
     посвя
    -0.32
    it
    -0.32
    sting
    -0.32
    gin
    -0.32
    le
    -0.32
    POSITIVE LOGITS
    WriteTagHelper
    0.59
     eventdata
    0.59
     defaultstate
    0.56
    PerformLayout
    0.54
    StoreMessageInfo
    0.53
     تضيفلها
    0.53
    intendent
    0.52
     évêque
    0.52
     ſol
    0.52
    majánló
    0.52
    Activations Density 0.270%

    No Known Activations