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. 14378
    Prev
    Next
    INDEX
    Explanations

    terms related to declarations, especially in legal or martial contexts

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

    declarations or announcements

    np_acts-logits-general · gemini-2.0-flash

    official declarations

    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
    Erreferentziak
    -0.68
     SPDX
    -0.54
    fjspx
    -0.54
    Hochspringen
    -0.53
     onBind
    -0.52
    Bronnen
    -0.51
     Glaub
    -0.51
    存于互联网档案馆
    -0.51
    KommentareTeilen
    -0.50
    Kanpo
    -0.48
    POSITIVE LOGITS
     declared
    1.52
     declare
    1.41
     declaration
    1.38
    declared
    1.30
     declaring
    1.29
     Declare
    1.19
     declares
    1.18
     emergency
    1.17
    Declare
    1.08
     declarar
    1.07
    Activations Density 0.171%

    No Known Activations