© 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 - 16k
    5. 18-GEMMASCOPE-RES-16K
    6. 3323
    Prev
    Next
    INDEX
    Explanations

    the repetition of the word "so" in various contexts

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

    Follows "so" or "mu"

    np_acts-logits-general · gemini-2.0-flash

    so much, so many, so far

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Sol
    -0.58
    Sol
    -0.57
    __
    -0.56
    m
    -0.54
    sol
    -0.51
     sol
    -0.50
    More
    -0.49
    p
    -0.49
    --
    -0.48
    n
    -0.48
    POSITIVE LOGITS
     much
    0.95
     propOrder
    0.92
     far
    0.86
     bezeichneter
    0.81
     Vikipedi
    0.80
     darn
    0.79
     мәкал
    0.76
    apimachinery
    0.75
    ckeye
    0.75
    othed
    0.75
    Activations Density 0.087%

    No Known Activations