© 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
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 14428
    Prev
    Next
    INDEX
    Explanations

    phrases expressing a moderate degree of quality or quantity

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

    phrases using a degree-intensifier adverb to add emphasis, especially in “X a …” or “X the …” constructions modifying adjectives or nouns.

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

    the word "quite" followed by an adjective or article.

    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
     atras
    -0.46
     return
    -0.46
     Cardona
    -0.40
    wendungs
    -0.39
     arac
    -0.36
     DRS
    -0.36
     Henderson
    -0.36
     recipients
    -0.35
    </em>
    -0.34
     LCCN
    -0.34
    POSITIVE LOGITS
    pretty
    1.00
     pretty
    0.99
    Pretty
    0.94
     fairly
    0.91
     Pretty
    0.91
     довольно
    0.87
     quite
    0.86
     Fairly
    0.85
     bastante
    0.85
     assez
    0.85
    Activations Density 0.090%

    No Known Activations