© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    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. 14393
    Prev
    Next
    INDEX
    Explanations

    terms and phrases related to revisions and changes in context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    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
     Promise
    -0.40
     voordeel
    -0.38
     voordelen
    -0.37
    好处
    -0.37
     lắm
    -0.36
     benefits
    -0.35
     garantiert
    -0.34
     Benefits
    -0.33
     führen
    -0.33
    FirebaseUser
    -0.33
    POSITIVE LOGITS
    altered
    0.70
    AlterField
    0.69
    
    0.68
    Changing
    0.67
     Altered
    0.66
     resourceCulture
    0.63
     Verſ
    0.63
     Changing
    0.62
     altering
    0.61
     CHANG
    0.61
    Activations Density 0.350%

    No Known Activations