© 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. 5484
    Prev
    Next
    INDEX
    Explanations

    descriptive adjectives that convey positive or beautiful qualities

    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
    makeConstraints
    -0.69
     greatness
    -0.65
    importance
    -0.63
    openConnection
    -0.62
     excellence
    -0.62
     dAtA
    -0.60
    MemoryWarning
    -0.60
    excellence
    -0.57
     otomatig
    -0.57
    rungsseite
    -0.57
    POSITIVE LOGITS
     looking
    0.57
    looking
    0.51
     smelling
    0.49
     sounding
    0.49
     thing
    0.48
     little
    0.47
     things
    0.46
    ties
    0.46
    Looking
    0.41
     crafted
    0.40
    Activations Density 0.247%

    No Known Activations