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. 2-GEMMASCOPE-RES-16K
    6. 6659
    Prev
    Next
    INDEX
    Explanations

    phrases or terms indicating significant amounts or levels of impact

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 2-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.2.hook_resid_post
    Hook Layer
    2
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     vzor
    -0.52
    utih
    -0.50
    
    -0.49
     stereotypes
    -0.48
     must
    -0.48
    ộp
    -0.48
    ed
    -0.48
    o
    -0.46
     vir
    -0.46
    paka
    -0.46
    POSITIVE LOGITS
     considerable
    2.03
    considerable
    1.88
     substantial
    1.80
     Substantial
    1.66
    substantial
    1.63
     Considerable
    1.62
    stantial
    1.53
     sizable
    1.42
     sizeable
    1.41
     appreciable
    1.30
    Activations Density 0.332%

    No Known Activations