Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 24-GEMMASCOPE-RES-16K
    6. 12435
    Prev
    Next
    INDEX
    Explanations

    references to the word "Bel" and its variations, indicating a focus on names or terms related to "Bel"

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

    starting with "Bel" or "bel"

    np_acts-logits-general · gemini-2.0-flash

    Bel- followed by words like voir, en, perio, ittles, gorod, cast

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     himo
    -0.74
    bootstrapcdn
    -0.71
     OFDb
    -0.70
    SourceChecksum
    -0.69
     ComVisible
    -0.59
    enumi
    -0.56
    دانشنامهٔ
    -0.56
    hysema
    -0.56
    richTextPanel
    -0.55
     اكتوبر
    -0.55
    POSITIVE LOGITS
     Bel
    0.55
     BEL
    0.53
     bel
    0.48
    canto
    0.47
    Bel
    0.47
    IntoConstraints
    0.46
    AppCompat
    0.46
    air
    0.46
     canto
    0.45
     AIR
    0.45
    Activations Density 0.117%

    No Known Activations