Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 16-GEMMASCOPE-RES-16K
    6. 6367
    Prev
    Next
    INDEX
    Explanations

    themes related to societal roles and discrimination

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

    Boycotts and ostracization

    np_acts-logits-general · gemini-2.0-flash

    shunned, ostracized, boycotted

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 16-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    LoggerFactory
    -0.54
    atigable
    -0.49
     Coroner
    -0.47
    žd
    -0.46
    原始内容存档于
    -0.46
    spyOn
    -0.44
    weak
    -0.44
     dilated
    -0.43
     LoggerFactory
    -0.43
    ایت
    -0.42
    POSITIVE LOGITS
    tonode
    0.68
     ostra
    0.67
     shunned
    0.66
    ########.
    0.63
     toegang
    0.60
     access
    0.60
     Wikimedijinoj
    0.60
    دانشنامهٔ
    0.60
    GEBURTSDATUM
    0.60
     boycot
    0.59
    Activations Density 0.413%

    No Known Activations