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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 3-GEMMASCOPE-RES-16K
    6. 16216
    Prev
    Next
    INDEX
    Explanations

    expressions related to offensiveness and inappropriate language

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.3.hook_resid_post
    Hook Layer
    3
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    __':
    
    -0.84
    saraba
    -0.77
    parsedMessage
    -0.76
     EconPapers
    -0.73
     WebDriverWait
    -0.72
     تضيفلها
    -0.68
    ]")]
    -0.68
    tagHelperRunner
    -0.66
     Савезне
    -0.65
    Искәрмәләр
    -0.65
    POSITIVE LOGITS
     offended
    1.60
     offend
    1.47
     offense
    1.38
     offence
    1.37
     offending
    1.35
     offensive
    1.23
     Offense
    1.14
    offensive
    1.13
     Offensive
    1.07
     affront
    1.07
    Activations Density 0.426%

    No Known Activations