© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 792
    Prev
    Next
    INDEX
    Explanations

    phrases related to legal defenses and justifications in actions

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

    hypothetical scenarios involving legal reasoning, driving mistakes, or consequences of actions.

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered by @neilrathi

    rhetorical analogies and similes that explain abstract points through concrete, illustrative scenarios.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408

    analogies and metaphorical comparisons (typically introduced by "like" or direct comparison structures).

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    Prompts (Dashboard)
    24,576 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
     ModelExpression
    -0.58
    ArrowToggle
    -0.47
    exitRule
    -0.45
    obald
    -0.38
    RISE
    -0.37
    standen
    -0.36
     predecessor
    -0.36
     GenerationType
    -0.36
    ceğ
    -0.35
    GOTREF
    -0.35
    POSITIVE LOGITS
     Италијани
    0.68
     EconPapers
    0.59
     мәкал
    0.56
     Comprometido
    0.51
     $_[
    0.49
     HasFactory
    0.48
     OnTrigger
    0.48
    AndEndTag
    0.47
    ThemeOverlay
    0.45
    dafx
    0.45
    Activations Density 0.198%

    No Known Activations