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. 20-GEMMASCOPE-RES-16K
    6. 1384
    Prev
    Next
    INDEX
    Explanations

    phrases related to problem-solving and decision-making challenges

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

    "Figure out", "got it", or similar phrases

    np_acts-logits-general · gemini-2.0-flash

    figure out

    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-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
    BeginContext
    -0.72
     saites
    -0.70
    ngdoc
    -0.64
    enumii
    -0.63
    ExtendWith
    -0.63
    adpleegd
    -0.62
    GraphicsUnit
    -0.61
    IndentedString
    -0.61
    Expecting
    -0.60
     rentrée
    -0.59
    POSITIVE LOGITS
     figured
    1.38
     figure
    1.08
    figure
    1.07
     figuring
    0.99
    figured
    0.97
     mastered
    0.92
     nailed
    0.88
     Figure
    0.86
     perfected
    0.85
    Figure
    0.85
    Activations Density 0.204%

    No Known Activations