© 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. 18-GEMMASCOPE-RES-16K
    6. 4492
    Prev
    Next
    INDEX
    Explanations

    references to research studies and their findings related to synaptic mechanisms

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

    Numbers in brackets

    np_acts-logits-general · gemini-2.0-flash

    citations and references

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    )/
    -0.57
    )-
    -0.41
    relle
    -0.39
     се
    -0.39
    stateProvider
    -0.39
    sl
    -0.37
    dart
    -0.37
    <bos>
    -0.36
    dataTable
    -0.35
    ętr
    -0.35
    POSITIVE LOGITS
    LayoutStyle
    0.83
    shafen
    0.67
    Хьажоргаш
    0.64
    twimg
    0.64
    SequentialGroup
    0.64
    NUMX
    0.62
    ValueStyle
    0.62
     chartered
    0.62
    .",
    
    0.62
    CrossRef
    0.62
    Activations Density 0.880%

    No Known Activations