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. 18-GEMMASCOPE-RES-16K
    6. 1876
    Prev
    Next
    INDEX
    Explanations

    patterns and functions related to regular expressions and matching strings

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

    Regular expressions with lookaheads or lookbehinds

    np_acts-logits-general · gemini-2.0-flash

    regular expression syntax

    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
    enumi
    -0.58
    تفصیلات
    -0.57
    )』
    -0.52
    Personendaten
    -0.46
    "]];
    -0.45
    ("*");
    -0.45
    以上就是
    -0.44
    RefNanny
    -0.44
    ',{
    -0.44
    unsaturated
    -0.44
    POSITIVE LOGITS
     vPvB
    0.65
     contextLoads
    0.62
    getMinutes
    0.60
    Rüyada
    0.59
     otomatig
    0.57
     BufferedReader
    0.56
    ULER
    0.56
    ^\
    0.56
     cdti
    0.56
    SupportActionBar
    0.56
    Activations Density 0.167%

    No Known Activations