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. 3639
    Prev
    Next
    INDEX
    Explanations

    instances indicating concerns about societal issues and inequalities

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

    following prepositions

    np_acts-logits-general · gemini-2.0-flash

    phrases with common starting words

    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
    polation
    -0.70
     
    -0.66
    nasium
    -0.63
    adecimal
    -0.62
    сутствие
    -0.56
    enment
    -0.54
    mær
    -0.54
    sterdam
    -0.53
    calaure
    -0.53
    riguez
    -0.52
    POSITIVE LOGITS
     disambiguazione
    0.96
    contentLoaded
    0.95
     myſelf
    0.94
     estekak
    0.87
     leaſt
    0.81
     purpoſe
    0.81
     Theſe
    0.80
     oprot
    0.79
     houſe
    0.78
    й
    0.78
    Activations Density 1.910%

    No Known Activations