Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 22-GEMMASCOPE-RES-16K
    6. 3379
    Prev
    Next
    INDEX
    Explanations

    keywords and phrases related to legal or investigative contexts

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

    Follows prepositions

    np_acts-logits-general · gemini-2.0-flash

    preposition followed by specific concepts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    MigrationBuilder
    -0.94
     resourceCulture
    -0.83
    errHandler
    -0.77
    enumi
    -0.74
    protoimpl
    -0.74
    bootstrapcdn
    -0.73
    complexContent
    -0.72
    tagHelperRunner
    -0.70
    Enllaces
    -0.69
     externi
    -0.69
    POSITIVE LOGITS
     The
    0.86
    quele
    0.66
    orexia
    0.65
     Of
    0.63
    wards
    0.60
    phans
    0.60
     Its
    0.59
    cestors
    0.59
    omalies
    0.58
     And
    0.56
    Activations Density 0.185%

    No Known Activations