© 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
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 15299
    Prev
    Next
    INDEX
    Explanations

    phrases or expressions related to wanting to learn or know more

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

    Followed by "more", "vi", "jte", or "better"

    np_acts-logits-general · gemini-2.0-flash

    know/learn more/better

    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
     tercio
    -0.46
    very
    -0.45
     
    -0.44
    mucho
    -0.43
     PLWABN
    -0.41
    Full
    -0.40
    much
    -0.40
    itz
    -0.39
    <eos>
    -0.39
    Final
    -0.38
    POSITIVE LOGITS
    RegressionTest
    1.25
    rungsseite
    1.06
    NUMX
    0.94
     ujednoznacz
    0.92
    expandindo
    0.92
     myſelf
    0.91
    曖昧さ回避
    0.87
    BibitemShut
    0.85
     ―――――
    0.83
    findpost
    0.83
    Activations Density 0.286%

    No Known Activations