© 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
    4. Residual Stream - 16k
    5. 33-GEMMASCOPE-RES-16K
    6. 697
    Prev
    Next
    INDEX
    Explanations

    phrases referring to individuals or entities and their actions or states

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 33-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_33/width_16k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.33.hook_resid_post
    Hook Layer
    33
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     itself
    -0.36
    itself
    -0.32
    also
    -0.32
     тоже
    -0.29
    some
    -0.28
    Squared
    -0.28
    many
    -0.28
    None
    -0.28
    seg
    -0.28
    I
    -0.28
    POSITIVE LOGITS
    DeleteBehavior
    0.68
    WithIOException
    0.64
    addCriterion
    0.63
    DotNetBar
    0.63
    prepareStatement
    0.62
    sizeCache
    0.61
    pecabe
    0.61
    HandlerContext
    0.61
     participaron
    0.58
     betweenstory
    0.58
    Activations Density 0.356%

    No Known Activations