© 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. 526
    Prev
    Next
    INDEX
    Explanations

    sections of text that introduce or detail "About" information referencing people, organizations, or projects

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

    introductory phrases or sections about facilities or personal information

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

    about followed by subject

    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
     ModelExpression
    -0.88
    errHandler
    -0.73
     jsPsych
    -0.72
    enterOuterAlt
    -0.72
    PreferredItem
    -0.72
    avadoc
    -0.69
     للمعارف
    -0.69
    aspectj
    -0.69
    Explicación
    -0.68
     SwitchCompat
    -0.67
    POSITIVE LOGITS
     Author
    0.62
     us
    0.62
     me
    0.62
     Us
    0.58
    press
    0.56
     Post
    0.55
     author
    0.55
     About
    0.54
     Me
    0.54
    Press
    0.53
    Activations Density 0.016%

    No Known Activations