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. 18-GEMMASCOPE-RES-16K
    6. 3708
    Prev
    Next
    INDEX
    Explanations

    comparative phrases highlighting quantity or degree

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

    Immediately preceded by "more"

    np_acts-logits-general · gemini-2.0-flash

    more than comparisons

    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
    bufio
    -0.58
     meanwhile
    -0.56
    Referencies
    -0.55
    endl
    -0.53
    Collegamenti
    -0.50
     ​​
    -0.49
    Various
    -0.48
     légales
    -0.48
    نجی
    -0.47
     ImageIcon
    -0.46
    POSITIVE LOGITS
     propOrder
    0.85
    MigrationBuilder
    0.83
     ExecuteAsync
    0.76
    #+#
    0.74
    Rüyada
    0.71
    neté
    0.67
    >>>>>>>>
    0.66
     just
    0.66
     doubling
    0.63
     conqueror
    0.63
    Activations Density 0.074%

    No Known Activations