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-2-9B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 310
    Prev
    Next
    INDEX
    Explanations

    various names and identifiers related to people and entities, particularly in a context involving products or contributions in certain industries

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

    names of people, particularly proper names.

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered by @neilrathi
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    Prompts (Dashboard)
    24,576 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
     didReceive
    -0.39
     whole
    -0.35
     inſ
    -0.34
     reaſon
    -0.34
     Holy
    -0.33
     dur
    -0.33
     ſever
    -0.32
    waitKey
    -0.32
    ]!=
    -0.32
     Barry
    -0.31
    POSITIVE LOGITS
     propOrder
    0.83
     Normdatei
    0.83
    ंदीखरीदारी
    0.81
     Мексичка
    0.77
    abestanden
    0.77
     Numerade
    0.76
    contentLoaded
    0.75
    OGND
    0.74
    SharedDtor
    0.73
     ModelExpression
    0.73
    Activations Density 2.506%

    No Known Activations