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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 20-GEMMASCOPE-RES-16K
    6. 12752
    Prev
    Next
    INDEX
    Explanations

    the repetitive use of the word "another" in various contexts

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

    the word "another"

    np_acts-logits-general · gemini-2.0-flash

    Introducing another item

    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
     autres
    -0.77
     تضيفلها
    -0.77
     lainnya
    -0.77
     demais
    -0.75
     others
    -0.70
     saites
    -0.70
     Roskov
    -0.68
    others
    -0.67
     către
    -0.67
     ovunque
    -0.65
    POSITIVE LOGITS
     layer
    0.76
     reason
    0.74
     couple
    0.71
     interesting
    0.70
     thing
    0.69
     version
    0.68
    worldly
    0.67
     chance
    0.67
     dozen
    0.67
     important
    0.66
    Activations Density 0.072%

    No Known Activations