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. 5-GEMMASCOPE-RES-16K
    6. 8695
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "interesting" and related descriptors in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.5.hook_resid_post
    Hook Layer
    5
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ClientRect
    -0.88
     étoient
    -0.78
     tervez
    -0.77
     silenzio
    -0.74
     kunjung
    -0.73
     giù
    -0.72
     mengucap
    -0.72
     fossero
    -0.71
     picioare
    -0.70
    ControllerAdvice
    -0.69
    POSITIVE LOGITS
     interesting
    2.26
    interesting
    2.19
     Interesting
    2.04
    Interesting
    1.99
     interessante
    1.63
     intéressante
    1.63
     interesante
    1.62
     fascinating
    1.61
     interessant
    1.55
     intéressant
    1.48
    Activations Density 0.073%

    No Known Activations