Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 11-GEMMASCOPE-RES-16K
    6. 6262
    Prev
    Next
    INDEX
    Explanations

    references to schemes or methodologies in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.11.hook_resid_post
    Hook Layer
    11
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Efq
    -0.85
     poffible
    -0.84
     poffe
    -0.77
     ſeveral
    -0.75
    »»
    -0.74
     EAT
    -0.73
     Overton
    -0.73
     myſelf
    -0.71
     faſt
    -0.71
     Diſ
    -0.70
    POSITIVE LOGITS
     SCHEME
    1.70
    Scheme
    1.65
     Scheme
    1.63
     Schemes
    1.61
     scheme
    1.61
     schemes
    1.59
    scheme
    1.58
    Schemes
    1.53
    cheme
    1.52
    SCHEME
    1.51
    Activations Density 0.057%

    No Known Activations