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 - 65k
    5. 20-GEMMASCOPE-RES-65K
    6. 1018
    Prev
    Next
    INDEX
    Explanations

    conditional statements and expressions of doubt or uncertainty

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

    conditional statements that express hypothetical situations

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

    not negated

    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-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    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
    assertNull
    -0.78
     propOrder
    -0.77
     Aucune
    -0.76
    impossible
    -0.75
    WebVitals
    -0.74
     impossible
    -0.74
     žiad
    -0.74
     žád
    -0.73
     ujednoznacz
    -0.73
    mybatisplus
    -0.73
    POSITIVE LOGITS
     not
    1.52
    not
    1.29
     isn
    1.26
     nicht
    1.25
     bukan
    1.23
     Not
    1.22
     aren
    1.19
    不是
    1.13
    Not
    1.11
     wasn
    1.07
    Activations Density 0.391%

    No Known Activations