© 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 - 16k
    5. 18-GEMMASCOPE-RES-16K
    6. 8424
    Prev
    Next
    INDEX
    Explanations

    address-related information and references

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

    Address-related terms

    np_acts-logits-general · gemini-2.0-flash

    street types and suite numbers

    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
    postIndex
    -0.74
    AllAfrica
    -0.60
    amoja
    -0.59
    requestCode
    -0.54
     stress
    -0.54
    saraba
    -0.54
    })$}
    -0.54
    VideoCapture
    -0.54
    --)
    
    -0.53
     geslacht
    -0.52
    POSITIVE LOGITS
     Suite
    0.86
    Suite
    0.70
     ExecuteAsync
    0.67
     SUITE
    0.66
     suite
    0.64
    suite
    0.61
     STE
    0.59
     Suites
    0.58
    Handlung
    0.54
     Apt
    0.54
    Activations Density 0.050%

    No Known Activations