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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 65k
    5. 16-GEMMASCOPE-RES-65K
    6. 28021
    Prev
    Next
    INDEX
    Explanations

    sequences of numerical values and their patterns

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

    mathematical expressions with powers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ########.
    -1.20
     فريبيس
    -1.20
     ErrIntOverflow
    -0.94
    uxxxx
    -0.92
    ఔ
    -0.91
    rungsseite
    -0.90
     يتيمه
    -0.84
     purpoſe
    -0.81
    WebVitals
    -0.81
    aarrggbb
    -0.80
    POSITIVE LOGITS
    FailureListener
    0.49
    makeConstraints
    0.47
     le
    0.42
    addFlags
    0.42
    lois
    0.41
    udan
    0.39
     **
    0.38
    mybatisplus
    0.38
     land
    0.38
    </h1>
    0.38
    Activations Density 0.305%

    No Known Activations