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. 22-GEMMASCOPE-RES-16K
    6. 5870
    Prev
    Next
    INDEX
    Explanations

    conditional phrases involving "either/or" constructs

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

    "Either" or "whether"

    np_acts-logits-general · gemini-2.0-flash

    presents alternatives or conditions

    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
     almeno
    -0.72
     nevertheless
    -0.64
     entrambi
    -0.63
     även
    -0.60
     withal
    -0.59
     Daarna
    -0.58
     همچنین
    -0.58
     presumably
    -0.58
     demás
    -0.58
     также
    -0.56
    POSITIVE LOGITS
    MemoryWarning
    0.70
     singly
    0.63
     directly
    0.60
    RectangleBorder
    0.60
     wholly
    0.59
     overtly
    0.57
     formal
    0.57
     незавершена
    0.57
    agisse
    0.57
     berupa
    0.56
    Activations Density 0.182%

    No Known Activations