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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 27-GEMMASCOPE-RES-16K
    11. 9083
    Prev
    Next
    INDEX
    Explanations

    instances of the verb "to be" in various forms, as well as related affirmations and descriptors

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 27-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_27/width_16k/average_l0_118
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.27.hook_resid_post
    Hook Layer
    27
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Demografie
    -0.43
     poffible
    -0.40
     isSet
    -0.39
     ragaz
    -0.38
     granat
    -0.36
    minecraftforge
    -0.36
     lenker
    -0.36
    twimg
    -0.35
     preen
    -0.34
    prefixer
    -0.34
    POSITIVE LOGITS
     willing
    0.58
     flattered
    0.55
     pretty
    0.54
     tired
    0.54
     afraid
    0.54
     gonna
    0.52
     incapable
    0.52
     very
    0.51
     going
    0.50
     stuck
    0.50
    Activations Density 0.434%

    No Known Activations