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. 21-GEMMASCOPE-RES-16K
    6. 7326
    Prev
    Next
    INDEX
    Explanations

    references to the word "Old" and its variations, indicating a focus on themes related to age or antiquity

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ViewFeatures
    -0.88
    Personensuche
    -0.78
    ManagedBean
    -0.74
     myſelf
    -0.74
     Ressources
    -0.71
    PyExc
    -0.69
     pleaſure
    -0.65
     itſelf
    -0.65
    OGND
    -0.65
    esgue
    -0.62
    POSITIVE LOGITS
     fashioned
    1.36
    fashioned
    1.11
    school
    0.83
     adage
    0.72
     school
    0.71
    timer
    0.70
     style
    0.69
     доб
    0.69
     age
    0.69
    style
    0.68
    Activations Density 0.090%

    No Known Activations