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. 4880
    Prev
    Next
    INDEX
    Explanations

    punctuation marks in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    Text after commas

    np_acts-logits-general · gemini-2.0-flash

    commas followed by conjunctions

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     Efq
    -1.20
     ―――――
    -1.19
     Majefty
    -1.15
     Anſ
    -1.14
     Houſe
    -1.08
    ſelves
    -1.08
    ſelf
    -1.08
     ་་
    -1.06
     ――――
    -1.00
     itſelf
    -0.99
    POSITIVE LOGITS
     it
    1.49
     we
    1.35
     they
    1.35
     there
    1.30
     I
    1.26
     you
    1.10
     so
    1.08
     but
    1.07
     this
    1.05
     It
    1.05
    Activations Density 0.533%

    No Known Activations