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. 4607
    Prev
    Next
    INDEX
    Explanations

    conditional phrases that imply requirements or necessities

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

    "as" or "if" followed by "necessary," "needed," or similar words

    np_acts-logits-general · gemini-2.0-flash

    as needed or if necessary

    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
     كومونز
    -0.66
    ChildIndex
    -0.62
     referenties
    -0.58
    InvalidProtocol
    -0.58
    UserScript
    -0.56
    kton
    -0.56
    UnknownFields
    -0.56
    曖昧さ回避
    -0.56
     resourceCulture
    -0.55
    horabuena
    -0.55
    POSITIVE LOGITS
     necessary
    1.84
     needed
    1.72
    necessary
    1.39
     required
    1.39
    needed
    1.35
     Necessary
    1.30
     appropriate
    1.29
     Needed
    1.22
     necesario
    1.18
    Necessary
    1.17
    Activations Density 0.267%

    No Known Activations