© 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. 1388
    Prev
    Next
    INDEX
    Explanations

    initialization and property settings in code

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

    structured programming syntax elements

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

    maximum values and related functions

    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
     disciple
    -0.47
     Bryan
    -0.46
    UNDER
    -0.43
    uden
    -0.42
    biet
    -0.42
    mol
    -0.41
     Christopher
    -0.40
     entry
    -0.40
     Ritter
    -0.40
     Sci
    -0.39
    POSITIVE LOGITS
    max
    1.84
     max
    1.81
     MAX
    1.67
    MAX
    1.64
     Max
    1.63
    Max
    1.62
     maximum
    1.57
    setMax
    1.52
     maxima
    1.51
     getMax
    1.50
    Activations Density 0.052%

    No Known Activations