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. 13077
    Prev
    Next
    INDEX
    Explanations

    variations of the word "une" and related terms, suggesting a focus on specific linguistic forms or spellings

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

    starting with "une", "pe", "ane", "ine", or "E"

    np_acts-logits-general · gemini-2.0-flash

    unequivocally, uneventful

    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
    CUIT
    -0.66
     Tears
    -0.63
     <<<<<<<<<<<<<<
    -0.60
    IUrlHelper
    -0.60
    segundo
    -0.58
    phrag
    -0.57
    setVerticalGroup
    -0.56
    Lugares
    -0.55
     FANDOM
    -0.55
    retty
    -0.54
    POSITIVE LOGITS
    ValueStyle
    0.59
    xt
    0.58
    ase
    0.56
    qui
    0.52
    qual
    0.50
    vent
    0.49
     xt
    0.49
     xc
    0.47
    xtra
    0.46
     élas
    0.45
    Activations Density 0.295%

    No Known Activations