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. 20-GEMMASCOPE-RES-16K
    6. 8895
    Prev
    Next
    INDEX
    Explanations

    references to free or spare time activities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ngrx
    -0.52
    abı
    -0.43
    Docs
    -0.42
    tonode
    -0.41
    FieldBuilder
    -0.41
    man
    -0.40
    essential
    -0.40
    oux
    -0.39
    ံ
    -0.39
    makeText
    -0.39
    POSITIVE LOGITS
     spare
    1.97
    spare
    1.71
     free
    1.58
     Spare
    1.53
     leisure
    1.46
    Spare
    1.46
    free
    1.41
    Free
    1.32
     свобод
    1.31
     downtime
    1.29
    Activations Density 0.184%

    No Known Activations