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. 1432
    Prev
    Next
    INDEX
    Explanations

    instances of placeholders or variables commonly used in programming and configuration files

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

    `APPLICATION_NAME` variable

    np_acts-logits-general · gemini-2.0-flash

    application name placeholder

    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
     "
    -0.59
     {"
    -0.51
     “
    -0.51
     «
    -0.48
     cry
    -0.48
     ["
    -0.45
     cum
    -0.45
     ``
    -0.44
     tra
    -0.44
     "");
    -0.43
    POSITIVE LOGITS
     Савезне
    0.74
    iNdEx
    0.70
    ngdoc
    0.66
     StringTokenizer
    0.63
    httphttps
    0.63
    AddTagHelper
    0.61
    numerusform
    0.61
    évaluateur
    0.61
     salut
    0.60
    كويكب
    0.60
    Activations Density 0.253%

    No Known Activations