© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 22-GEMMASCOPE-RES-16K
    6. 14237
    Prev
    Next
    INDEX
    Explanations

    the presence of regulatory or legal changes and their implications

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

    exceptions or exemptions

    np_acts-logits-general · gemini-2.0-flash

    exemption and relief

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    峙
    -0.43
     στι
    -0.38
    INSEE
    -0.36
     Aseg
    -0.36
     Kontrolle
    -0.36
     proprement
    -0.36
    Autoritní
    -0.36
    تمع
    -0.35
     controllo
    -0.35
    CodedInputStream
    -0.35
    POSITIVE LOGITS
     exemption
    2.00
     exemptions
    1.98
     exceptions
    1.94
     exception
    1.78
     exempt
    1.68
     exempted
    1.66
     Exemption
    1.60
     Exceptions
    1.57
    exceptions
    1.43
    exempt
    1.37
    Activations Density 0.721%

    No Known Activations