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. 20-GEMMASCOPE-RES-16K
    6. 12769
    Prev
    Next
    INDEX
    Explanations

    mentions of members or groups associated with specific entities or organizations

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

    Following "of" or "in."

    np_acts-logits-general · gemini-2.0-flash

    members of groups

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ]")]
    -0.90
     ninguém
    -0.67
    مصادر
    -0.60
    posedge
    -0.59
    }}}}
    -0.59
     anybody
    -0.59
     itſelf
    -0.59
     alguém
    -0.59
    eseorang
    -0.58
    ).]
    -0.58
    POSITIVE LOGITS
     whom
    0.78
     who
    0.73
     audience
    0.64
    تفصیلات
    0.62
    whom
    0.57
    ترم
    0.56
     coro
    0.55
     know
    0.54
     sexe
    0.54
     entourage
    0.54
    Activations Density 0.387%

    No Known Activations