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. 3746
    Prev
    Next
    INDEX
    Explanations

    words related to various forms of "expensive" and "expatriation."

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

    Texts containing the abbreviation "Exp"

    np_acts-logits-general · gemini-2.0-flash

    explaining, expatriates, expulsion, expands

    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
    findpost
    -0.79
     AspNetCore
    -0.61
     <<<<<<<<<<<<<<
    -0.61
    RegistryLite
    -0.58
     CreateTagHelper
    -0.56
    TintMode
    -0.54
     sirena
    -0.54
    >*/
    -0.53
    ]='\
    -0.52
     Normdatei
    -0.52
    POSITIVE LOGITS
     EXP
    0.81
     exp
    0.75
     raiſ
    0.74
     EXPL
    0.73
    ฐ
    0.69
     Exp
    0.68
     fubject
    0.67
    EXP
    0.67
     eksp
    0.66
     themſelves
    0.65
    Activations Density 0.166%

    No Known Activations