© 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-2-2B
    4. Residual Stream - 65k
    5. 20-GEMMASCOPE-RES-65K
    6. 1551
    Prev
    Next
    INDEX
    Explanations

    the presence of XML declaration or instructions

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

    XML declaration tags

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

    question marks and special characters

    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-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    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
     autorytatywna
    -1.22
     дописавши
    -1.21
    Personensuche
    -1.14
     tartalomajánló
    -1.06
     invokingState
    -1.04
    +#+#
    -1.03
    rungsseite
    -1.03
    DeleteBehavior
    -1.00
    évaluateur
    -0.98
    Brainz
    -0.91
    POSITIVE LOGITS
    ?
    0.51
    true
    0.45
     Germ
    0.45
    Germ
    0.45
    pep
    0.45
    кра
    0.45
     Desmond
    0.43
    ди
    0.42
    δι
    0.41
     doublet
    0.41
    Activations Density 0.012%

    No Known Activations