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. 13606
    Prev
    Next
    INDEX
    Explanations

    occurrences of phrases indicating the first instance or time of an event

    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
    yticks
    -0.43
     for
    -0.42
    <bos>
    -0.40
    hen
    -0.39
     P
    -0.39
     sol
    -0.39
    olars
    -0.39
    δου
    -0.39
    länd
    -0.39
    后期
    -0.38
    POSITIVE LOGITS
     first
    1.97
    first
    1.65
     first
    1.46
     pertama
    1.44
     primera
    1.38
     första
    1.38
     firstly
    1.33
     primeira
    1.33
     første
    1.29
     впервые
    1.26
    Activations Density 0.080%

    No Known Activations