Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit 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. 7331
    Prev
    Next
    INDEX
    Explanations

    mentions of relationships

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

    "relationship"

    np_acts-logits-general · gemini-2.0-flash

    relationship with or between

    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
     ſmall
    -0.70
    TestingModule
    -0.70
    rceil
    -0.69
     whoſe
    -0.66
     Collected
    -0.66
     becauſe
    -0.65
     pngtree
    -0.65
     unreleased
    -0.64
     uſed
    -0.64
    ]<<"
    -0.63
    POSITIVE LOGITS
     relationships
    0.77
    Relationships
    0.69
     status
    0.68
     Beziehungen
    0.59
     Relationships
    0.59
     relationship
    0.58
    Ship
    0.56
    status
    0.55
    맺
    0.53
    building
    0.52
    Activations Density 0.174%

    No Known Activations