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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 12-GEMMASCOPE-RES-16K
    6. 1501
    Prev
    Next
    INDEX
    Explanations

    moments of surprise or unexpected events, particularly those that are positive or celebratory in nature.

    oai_token-act-pair · claude-3-5-sonnet-20240620Triggered by @bot

    elements and phrases related to special events and surprises

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.12.hook_resid_post
    Hook Layer
    12
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     consultato
    -0.80
     nakalista
    -0.67
     onOptions
    -0.66
     <",
    -0.65
    LLocation
    -0.60
     isNameExpr
    -0.59
     MotionEvent
    -0.59
    Notae
    -0.59
    NewUrlParser
    -0.57
     {}));
    -0.56
    POSITIVE LOGITS
     Surprise
    0.70
     surprise
    0.68
    юрпри
    0.64
    Surprise
    0.64
     surprises
    0.58
     ambush
    0.57
    surprise
    0.56
    istici
    0.54
     Customized
    0.52
     sorpresa
    0.51
    Activations Density 0.140%

    No Known Activations