© 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-2-2B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 10657
    Prev
    Next
    INDEX
    Explanations

    modal verbs indicating requests or inquiries

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

    "Can" followed by a question

    np_acts-logits-general · gemini-2.0-flash

    can followed by pronoun

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
    yntaxException
    -1.05
    تفصیلات
    -0.97
     Infórmanos
    -0.94
    évaluateur
    -0.94
     Roskov
    -0.91
     <<<<<<<<<<<<<<
    -0.91
    Diwedd
    -0.91
    ArrowToggle
    -0.88
    noopener
    -0.86
    EndContext
    -0.85
    POSITIVE LOGITS
     anyone
    1.10
     we
    1.05
     they
    1.04
     there
    0.98
     anybody
    0.97
     you
    0.92
    anyone
    0.81
     it
    0.80
     someone
    0.79
    Anyone
    0.75
    Activations Density 0.213%

    No Known Activations