© 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
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 16k
    9. 18-GEMMASCOPE-RES-16K
    11. 3380
    Prev
    Next
    INDEX
    Explanations

    references to methodologies or frameworks for data analysis and evaluation

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

    "Based on" and subsequent information

    np_acts-logits-general · gemini-2.0-flash

    based on observations

    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
    JvmStatic
    -0.65
     AssemblyTitle
    -0.59
    WriteBarrier
    -0.55
    InjectAttribute
    -0.54
    disposing
    -0.54
    頃は
    -0.53
     digress
    -0.51
     Extragalactic
    -0.51
    Autores
    -0.50
    mtable
    -0.49
    POSITIVE LOGITS
     observations
    0.69
     tilgjenge
    0.68
     principles
    0.66
     recent
    0.65
     available
    0.65
     knowledge
    0.65
     criteria
    0.64
     information
    0.63
    %</
    0.63
     previous
    0.62
    Activations Density 0.667%

    No Known Activations