© 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. 1474
    Prev
    Next
    INDEX
    Explanations

    programming terminologies related to artificial intelligence and creatures

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

    elements and actions related to programming or game mechanics, specifically in a creature or entity context

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

    select, get, GetData

    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
    ValueStyle
    -0.86
     оригіналу
    -0.70
    WriteTagHelper
    -0.55
    MigrationBuilder
    -0.53
    httphttps
    -0.52
    AsStream
    -0.47
    multer
    -0.47
    JSONException
    -0.47
    нодоро
    -0.46
    hlin
    -0.45
    POSITIVE LOGITS
     wearer
    0.80
     Efq
    0.78
     houſe
    0.68
     teammate
    0.67
     occupants
    0.63
     Nearby
    0.63
     nearby
    0.63
     poffe
    0.62
    ſelf
    0.61
    ReferenceEquals
    0.61
    Activations Density 0.077%

    No Known Activations