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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    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. 9193
    Prev
    Next
    INDEX
    Explanations

    strings that represent identifiers or keys, particularly in code or data structures

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

    Code snippets referencing "ngx"

    np_acts-logits-general · gemini-2.0-flash

    ngx and pixel drawing

    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
    ,
    -0.88
     ‘
    -0.83
     «
    -0.76
    /');
    -0.72
     et
    -0.69
     Fig
    -0.68
    .
    -0.68
    ]');
    -0.66
     so
    -0.66
    /'.$
    -0.66
    POSITIVE LOGITS
     pleaſure
    1.26
     reaſon
    1.23
    //"
    1.21
     purpoſe
    1.17
     myſelf
    1.16
     quæ
    1.14
     barbati
    1.12
    ]=="
    1.11
     juſt
    1.10
     itſelf
    1.10
    Activations Density 0.339%

    No Known Activations