© 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
    7. Residual Stream - 16k
    9. 31-GEMMASCOPE-RES-16K
    11. 3532
    Prev
    Next
    INDEX
    Explanations

    mentions of "Li" or its variations and associations in various contexts

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

    The neuron detects mentions of "Li" as a name or chemical element, often appearing at the start of last names or chemical compounds.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    the author name "Li" in academic citations and references.

    oai_token-act-pair · claude-4-5-haikuTriggered by @emiglarou
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 31-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_31/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.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
    ed
    -0.39
    Шо
    -0.37
    SafeAreaView
    -0.35
    werth
    -0.34
     hered
    -0.34
     ton
    -0.34
    arki
    -0.33
     Macle
    -0.33
     langer
    -0.33
    ACKS
    -0.33
    POSITIVE LOGITS
    setVerticalGroup
    0.78
     doInBackground
    0.77
    Personendaten
    0.69
    AndEndTag
    0.65
    الإنجليزية
    0.63
    extAlignment
    0.60
    AddTagHelper
    0.59
     betweenstory
    0.59
    MLLoader
    0.58
    EDEFAULT
    0.58
    Activations Density 0.262%

    No Known Activations