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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 21-GEMMASCOPE-RES-16K
    6. 2060
    Prev
    Next
    INDEX
    Explanations

    structured notation and special characters commonly used in mathematical or logical expressions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ↵↵
    -0.62
    <eos>
    -0.56
                                   
    -0.54
    atsch
    -0.50
    |$.
    -0.50
    %@",
    -0.47
    ियाँ
    -0.47
                                 
    -0.47
                                  
    -0.46
    不得
    -0.46
    POSITIVE LOGITS
    setcounter
    1.07
    abestanden
    0.96
    rungsseite
    0.92
    enumi
    0.92
    GEBURTS
    0.87
    LLocation
    0.86
     referrerpolicy
    0.84
    UserScript
    0.84
    TemporalType
    0.79
    AddTagHelper
    0.77
    Activations Density 0.007%

    No Known Activations