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. 22-GEMMASCOPE-RES-16K
    6. 10279
    Prev
    Next
    INDEX
    Explanations

    code-related structures, specifically class definitions and related annotations

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

    Code, Java, or Class-related terms

    np_acts-logits-general · gemini-2.0-flash

    Java class and method names

    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
    homonymie
    -0.53
     ujednoznacz
    -0.51
    nP
    -0.48
    arXiv
    -0.48
    retas
    -0.45
    EXTERN
    -0.45
    ✭✭
    -0.44
    वाह
    -0.43
    脚
    -0.42
    styleUrls
    -0.41
    POSITIVE LOGITS
    Personensuche
    0.68
    AccessFile
    0.65
    jdk
    0.65
    JDK
    0.65
     JVM
    0.63
    JVM
    0.63
     Java
    0.62
     Numerade
    0.61
     JDK
    0.61
     sun
    0.60
    Activations Density 0.439%

    No Known Activations