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. 972
    Prev
    Next
    INDEX
    Explanations

    terms related to funding and budgeting in research contexts

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

    phrases related to contractual terms and conditions

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

    construction contracts

    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
    ResumeLayout
    -0.73
    AutoScaleMode
    -0.71
    StructEnd
    -0.68
    __*/
    -0.67
    ActivityCompat
    -0.66
     nahilalakip
    -0.66
    UseVisualStyle
    -0.66
    zzá
    -0.64
     Référence
    -0.62
     localObject
    -0.62
    POSITIVE LOGITS
     project
    0.73
     contract
    0.73
     Contract
    0.73
     Construction
    0.69
     Project
    0.67
    Construction
    0.67
     subcontractors
    0.67
    contract
    0.67
     proyek
    0.66
    Contract
    0.66
    Activations Density 0.188%

    No Known Activations