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. 5-GEMMASCOPE-RES-16K
    6. 4130
    Prev
    Next
    INDEX
    Explanations

    greetings and conversational interjections

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.5.hook_resid_post
    Hook Layer
    5
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    scriptcase
    -0.76
     useRouter
    -0.68
     coa
    -0.68
     Rivière
    -0.66
     Brasileiro
    -0.66
    AfterClass
    -0.65
    oficina
    -0.64
    Cordialement
    -0.64
     trebui
    -0.63
    ówczas
    -0.63
    POSITIVE LOGITS
     Hey
    2.20
    Hey
    2.19
     hey
    2.09
     HEY
    1.96
    hey
    1.84
    HEY
    1.74
     Heywood
    1.35
    Heya
    1.27
    嘿
    1.11
     Hej
    1.10
    Activations Density 0.028%

    No Known Activations