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 - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 2323
    Prev
    Next
    INDEX
    Explanations

    themes of relatability and personal connection within the context of experiences and emotions

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

    connections and relatability

    np_acts-logits-general · gemini-2.0-flash

    relate and connect

    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-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 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
    ########.
    -0.81
     kasarigan
    -0.76
    </tfoot>
    -0.64
     שוליים
    -0.60
    Systematic
    -0.58
     ligiloj
    -0.57
    ⓧ
    -0.57
     betweenstory
    -0.56
    ."));
    -0.55
     errno
    -0.54
    POSITIVE LOGITS
     relate
    1.48
     resonate
    1.45
     relatable
    1.39
     empathi
    1.33
     resonates
    1.32
     relat
    1.27
     relates
    1.21
     connection
    1.20
     empathy
    1.16
     connect
    1.14
    Activations Density 0.205%

    No Known Activations