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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 22-GEMMASCOPE-RES-16K
    6. 10346
    Prev
    Next
    INDEX
    Explanations

    elements and operations associated with mathematical expressions and programming syntax

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

    Followed by "s" (or similar letter)

    np_acts-logits-general · gemini-2.0-flash

    s-variations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     NSCoder
    -0.70
    rungsseite
    -0.70
     utafitiHapana
    -0.69
    ]")]
    -0.62
    TagMode
    -0.59
     חיצוניים
    -0.58
     برانيه
    -0.58
    脚注の使い方
    -0.58
     GoogleFonts
    -0.56
     ligiloj
    -0.54
    POSITIVE LOGITS
     s
    2.39
    s
    1.49
     S
    1.22
    檚
    1.05
    ́s
    1.04
    s
    1.02
     س
    1.01
    S
    0.99
    𝑠
    0.97
     с
    0.85
    Activations Density 0.261%

    No Known Activations