© 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. 617
    Prev
    Next
    INDEX
    Explanations

    phrases indicating validation checks and conditions regarding data integrity

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

    expressions indicating conditions regarding data retrieval or API usage

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

    empty or null

    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
    äß
    -0.54
     Etsy
    -0.51
    rawdę
    -0.49
    Dés
    -0.49
    IsVisible
    -0.48
    carat
    -0.47
    tze
    -0.47
     etsy
    -0.47
    lus
    -0.46
     XK
    -0.46
    POSITIVE LOGITS
     null
    1.03
     empty
    0.95
     NULL
    0.94
    Null
    0.93
     Null
    0.92
    null
    0.90
    NULL
    0.87
    空
    0.82
    empty
    0.81
    NullException
    0.78
    Activations Density 0.072%

    No Known Activations