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. 3096
    Prev
    Next
    INDEX
    Explanations

    mentions of language and its various forms

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

    "language"

    np_acts-logits-general · gemini-2.0-flash

    language, languages,

    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
    ücks
    -0.71
    TypeDef
    -0.58
    ︎
    -0.57
     greateſt
    -0.56
    ſelves
    -0.55
    ſelf
    -0.55
     occafion
    -0.55
     reaſon
    -0.54
     preſent
    -0.54
     microm
    -0.54
    POSITIVE LOGITS
     Lang
    0.89
     LANG
    0.87
     Language
    0.78
    Languages
    0.73
     languages
    0.73
     Langu
    0.73
    Lang
    0.71
     язы
    0.71
     Languages
    0.70
    auge
    0.70
    Activations Density 0.106%

    No Known Activations