© 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. 20-GEMMASCOPE-RES-131K
    6. 122049
    Prev
    Next
    INDEX
    Explanations

    phrases related to asking questions and seeking closure

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

    countering opposition

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

    the neuron fires on phrasal/reciprocal actions marked by the particle "back" (e.g., "text back", "email him back", "give back").

    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
    WillAppear
    -0.39
     đây
    -0.37
    Linki
    -0.36
     rơi
    -0.35
    真好
    -0.35
     @"/
    -0.34
     Passing
    -0.34
     становника
    -0.34
     Presentations
    -0.34
     ảo
    -0.34
    POSITIVE LOGITS
     counter
    0.58
    AndEndTag
    0.53
     COUNTER
    0.50
     Counter
    0.50
     retali
    0.49
    counter
    0.47
     matching
    0.47
     Gleich
    0.46
     countered
    0.46
     counters
    0.45
    Activations Density 0.106%

    No Known Activations