© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 13263
    Prev
    Next
    INDEX
    Explanations

    phrases discussing noteworthy aspects of a performance or artistic presence.

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

    phrases related to physical appearance and personal attributes.

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot

    phrases or descriptions that convey a sense of intensity, complexity, or sophistication. It seems to activate strongly for words and phrases that express nuanced or multi-faceted qualities, especially in descriptions of art, performances, or complex ideas.

    oai_token-act-pair · claude-3-5-sonnet-20240620Triggered by @bot

    Word after adjectives ending in '-al'

    np_acts-logits-general · gemini-2.0-flash

    qualities and abstract concepts

    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
    windowFixed
    -0.67
     UIFont
    -0.67
     aján
    -0.63
    setVerticalGroup
    -0.61
     législation
    -0.58
    증
    -0.54
    FTFY
    -0.54
    cogni
    -0.54
     raiſ
    -0.54
    offsetof
    -0.54
    POSITIVE LOGITS
     simplicity
    1.06
     flexibility
    0.99
     elegance
    0.86
     complexity
    0.86
     robustness
    0.84
     purity
    0.83
     ease
    0.82
     versatility
    0.82
     reliability
    0.81
     rigidity
    0.81
    Activations Density 0.731%

    No Known Activations