© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 16k
    9. 16-GEMMASCOPE-RES-16K
    11. 9205
    Prev
    Next
    INDEX
    Explanations

    legal terms and phrases related to court decisions and legal arguments

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

    Follows legal citations or court decisions

    np_acts-logits-general · gemini-2.0-flash

    court legal decisions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 16-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    cedures
    -0.65
    horabuena
    -0.65
    Herzliche
    -0.61
    angering
    -0.61
    ratulations
    -0.61
    =$((
    -0.59
     iſt
    -0.58
    duled
    -0.58
     sessionFactory
    -0.57
     Мексичка
    -0.57
    POSITIVE LOGITS
     court
    0.56
     Court
    0.52
     CONSIDER
    0.51
    fromnode
    0.48
     Had
    0.48
     Magistrate
    0.47
     argument
    0.46
     consideramos
    0.46
     Judge
    0.45
     JUDGE
    0.45
    Activations Density 2.152%

    No Known Activations