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

    Neuronpedia

    Jacobian LensNEW
    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. 35-GEMMASCOPE-RES-16K
    11. 5795
    Prev
    Next
    INDEX
    Explanations

    references to specific types of sandwiches, particularly the French dip sandwich

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 35-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_35/width_16k/average_l0_120
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.35.hook_resid_post
    Hook Layer
    35
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    addCriterion
    -0.63
    AndEndTag
    -0.63
     gethan
    -0.60
    principalTable
    -0.58
     disambiguazione
    -0.53
     wuß
    -0.52
    httphttps
    -0.50
     ddelweddau
    -0.50
     unknownFields
    -0.49
     mußten
    -0.48
    POSITIVE LOGITS
     sandwich
    1.20
     Sandwich
    1.08
     sandwiches
    0.99
    sandwich
    0.98
    Sandwich
    0.98
     sand
    0.93
     burger
    0.90
     SAND
    0.88
     Sandwiches
    0.82
     Sand
    0.79
    Activations Density 0.196%

    No Known Activations