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. 25-GEMMASCOPE-RES-16K
    6. 5996
    Prev
    Next
    INDEX
    Explanations

    actions involved in creating, publishing, and engaging with content or initiatives

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 25-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_25/width_16k/average_l0_116
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.25.hook_resid_post
    Hook Layer
    25
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Any
    -0.98
     Any
    -0.89
     apapun
    -0.86
     qualquer
    -0.80
    any
    -0.80
    Anything
    -0.78
     Anything
    -0.77
    Either
    -0.77
     qualsiasi
    -0.76
     ANY
    -0.76
    POSITIVE LOGITS
     several
    2.19
    several
    1.84
     numerous
    1.84
     many
    1.61
     two
    1.54
    Several
    1.43
     quite
    1.42
     various
    1.41
     plusieurs
    1.40
     varios
    1.39
    Activations Density 0.848%

    No Known Activations