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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 37-GEMMASCOPE-RES-16K
    6. 993
    Prev
    Next
    INDEX
    Explanations

    conversational phrases and expressions indicating advice or personal experiences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 37-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_37/width_16k/average_l0_124
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.37.hook_resid_post
    Hook Layer
    37
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     contextLoads
    -0.50
     OMITBAD
    -0.49
    writeFieldEnd
    -0.48
     MainAxisSize
    -0.44
    ArgsConstructor
    -0.44
    
    -0.43
    niająca
    -0.40
    点此举报
    -0.40
    $_['
    -0.39
    \{\\
    -0.38
    POSITIVE LOGITS
     googling
    1.98
     google
    1.89
     researching
    1.85
     Google
    1.84
     research
    1.83
     googled
    1.80
     researched
    1.73
    Google
    1.65
    research
    1.60
     search
    1.58
    Activations Density 0.741%

    No Known Activations