© 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. 28165
    Prev
    Next
    INDEX
    Explanations

    references to emotional and psychological themes related to personal experiences and societal issues

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

    The neuron activates whenever it encounters prominent content nouns—especially abstract concepts, technical terms, or proper names that function as the head words of descriptive noun phrases.

    oai_token-act-pair · o3Triggered by @adamkarvonen

    end of clauses

    np_acts-logits-general · gemini-2.5-flash-lite

    the neuron detects salient content words—notably important nouns and intense descriptive phrases that signal the main topic or dramatic/emphatic concepts.

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    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
    <unused28>
    -1.57
    <unused8>
    -1.57
    <unused79>
    -1.57
    <unused80>
    -1.57
    <unused41>
    -1.57
    <unused74>
    -1.57
    <unused3>
    -1.56
    <unused14>
    -1.56
    [@BOS@]
    -1.56
    <pad>
    -1.56
    POSITIVE LOGITS
    .
    1.11
    ;
    0.58
    ,
    0.58
    !
    0.49
    ..
    0.48
    。
    0.48
     because
    0.47
    .,
    0.46
     after
    0.45
     with
    0.44
    Activations Density 9.935%

    No Known Activations