© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 16k
    9. 22-GEMMASCOPE-RES-16K
    11. 5622
    Prev
    Next
    INDEX
    Explanations

    programming terminology and concepts related to structured data handling and conditionals

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

    Follows tokens "TAG", "get", or "dot"

    np_acts-logits-general · gemini-2.0-flash

    `__` after identifiers

    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
     PDA
    -0.52
     ErrIntOverflow
    -0.51
     Koy
    -0.50
    nale
    -0.49
    Liang
    -0.48
     Crane
    -0.47
     TestBed
    -0.45
    PDA
    -0.45
     Liang
    -0.45
    Grat
    -0.44
    POSITIVE LOGITS
    __
    2.16
     __
    2.00
    .__
    1.66
    (__
    1.60
    )__
    1.52
     (__
    1.50
    &__
    1.39
     "__
    1.37
     '__
    1.35
    "__
    1.32
    Activations Density 0.295%

    No Known Activations