© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 65k
    5. 16-GEMMASCOPE-RES-65K
    6. 27906
    Prev
    Next
    INDEX
    Explanations

    inquiries related to understanding meaning or definitions

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

    or meaning quantify

    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
     quæ
    -0.80
     avoient
    -0.63
     يتيمه
    -0.61
    `{.
    -0.61
    DATABASES
    -0.61
    PyTuple
    -0.60
     (\<
    -0.60
     doubtnut
    -0.60
     Theſe
    -0.59
     againſt
    -0.58
    POSITIVE LOGITS
     mean
    1.05
     MEAN
    1.02
     meaning
    1.01
     Mean
    0.93
    mean
    0.91
    meaning
    0.85
     meanings
    0.82
    MEAN
    0.81
    Mean
    0.80
     Meaning
    0.80
    Activations Density 0.023%

    No Known Activations