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. 10-GEMMASCOPE-RES-16K
    6. 16275
    Prev
    Next
    INDEX
    Explanations

    punctuation and formatting marks typically used in citations or legal documents

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     
    -0.88
     K
    -0.77
     time
    -0.74
     '
    -0.72
     "
    -0.72
     F
    -0.72
     Time
    -0.70
    djangoproject
    -0.69
     Bra
    -0.69
     G
    -0.68
    POSITIVE LOGITS
     ainfi
    1.46
     Monfieur
    1.37
     avoient
    1.28
     étoient
    1.26
     plufieurs
    1.25
     Cæsar
    1.23
     enfans
    1.21
    Portale
    1.18
     feroit
    1.17
     myſelf
    1.16
    Activations Density 0.120%

    No Known Activations