© 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. 18-GEMMASCOPE-RES-16K
    11. 9736
    Prev
    Next
    INDEX
    Explanations

    contractions and possessive forms of pronouns

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

    Followed by "s" (or other letters), after apostrophe

    np_acts-logits-general · gemini-2.0-flash

    interjections and filler words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     незавершена
    -0.90
    endpush
    -0.69
    fjspx
    -0.65
    PhysRev
    -0.64
    EDEFAULT
    -0.59
    ViewFeatures
    -0.57
    __':
    -0.57
    nodoc
    -0.57
     EconPapers
    -0.56
    —“
    -0.55
    POSITIVE LOGITS
    '
    0.62
     um
    0.60
     ball
    0.58
     uh
    0.56
     Uh
    0.55
     Um
    0.55
     VIDEOTAPE
    0.53
     Okay
    0.51
    RegressionTest
    0.51
     okay
    0.50
    Activations Density 0.067%

    No Known Activations