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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 18-GEMMASCOPE-RES-16K
    6. 308
    Prev
    Next
    INDEX
    Explanations

    specific words and patterns related to achievements or notable events, especially in academic or formal contexts

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

    Tokens starting with "ern"

    np_acts-logits-general · gemini-2.0-flash

    ern

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ViewFeatures
    -0.76
    +#+#
    -0.76
     externi
    -0.65
     ControllerBase
    -0.65
    parsedMessage
    -0.63
    \{\\
    -0.63
    évaluateur
    -0.62
    withstanding
    -0.62
    SaveChangesAsync
    -0.61
     bezeichneter
    -0.58
    POSITIVE LOGITS
    PreferredItem
    0.62
    ark
    0.60
    ars
    0.58
    urn
    0.58
     Erm
    0.54
     Tern
    0.53
     Turkmenistan
    0.52
    ORN
    0.52
    arn
    0.52
    alk
    0.51
    Activations Density 0.443%

    No Known Activations