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. 8606
    Prev
    Next
    INDEX
    Explanations

    instances of whitespace character(s)

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

    Capitalized words not at the beginning of sentences

    np_acts-logits-general · gemini-2.0-flash

    However, if

    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
    contentLoaded
    -0.93
     AssemblyVersion
    -0.89
     sánh
    -0.79
     propOrder
    -0.77
     CreateTagHelper
    -0.77
    DOCTYPE
    -0.76
    Bioaccumulative
    -0.75
    InjectAttribute
    -0.72
    stdc
    -0.71
    adaptiveStyles
    -0.70
    POSITIVE LOGITS
     itſelf
    1.09
     myſelf
    0.94
     Anſ
    0.87
     Monfieur
    0.85
     uſed
    0.84
     Jefus
    0.84
     ſeveral
    0.83
     Eſ
    0.83
     Reſ
    0.79
    The
    0.79
    Activations Density 0.102%

    No Known Activations