© 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. 6218
    Prev
    Next
    INDEX
    Explanations

    terms related to scientific or technical analysis

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

    Code or reference snippets

    np_acts-logits-general · gemini-2.0-flash

    sequences after 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
     nakalista
    -0.74
    AnchorStyles
    -0.72
    uguetes
    -0.71
     Majefty
    -0.70
    addContainerGap
    -0.69
     loch
    -0.69
     DRAM
    -0.68
     JIT
    -0.67
     SqlDbType
    -0.67
    getSystemService
    -0.66
    POSITIVE LOGITS
    @
    0.56
    Przypisy
    0.51
     chez
    0.50
     Meksiku
    0.48
    InitVars
    0.47
     Microsoft
    0.46
     @
    0.46
     ;$
    0.45
    member
    0.44
    《
    0.44
    Activations Density 1.779%

    No Known Activations