© 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 - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 10602
    Prev
    Next
    INDEX
    Explanations

    terms related to API sessions, identity verification, and configuration parameters

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

    Comes before colons, dashes, or quotes

    np_acts-logits-general · gemini-2.0-flash

    schedule policy or attributes

    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
    ,
    -0.65
    ↵
    -0.61
    ?
    -0.57
     
    -0.57
     +
    -0.57
    twimg
    -0.53
    Jeografia
    -0.52
      
    -0.52
     RequestMethod
    -0.49
    fara
    -0.49
    POSITIVE LOGITS
     Normdatei
    0.91
     الحره
    0.90
    extAlignment
    0.79
     ſind
    0.68
    theless
    0.67
     XG
    0.66
     doubtnut
    0.65
    TagHelper
    0.65
     stanovnika
    0.64
    वरण
    0.63
    Activations Density 0.473%

    No Known Activations