Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 14566
    Prev
    Next
    INDEX
    Explanations

    structured programming and testing constructs

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

    Code with parentheses, brackets, or equal signs

    np_acts-logits-general · gemini-2.0-flash

    equality checks

    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
    BeginContext
    -0.62
     незавершена
    -0.57
     Amen
    -0.55
    Abp
    -0.54
    .[/
    -0.53
    :%.
    -0.52
     Blon
    -0.52
    <%
    -0.52
    '):
    
    -0.52
     short
    -0.51
    POSITIVE LOGITS
     matches
    0.66
     exactly
    0.62
     stimmen
    0.61
    scalatest
    0.61
     ==
    0.60
     equalled
    0.59
     match
    0.59
     Exactly
    0.59
     exactement
    0.59
     fotografico
    0.58
    Activations Density 0.596%

    No Known Activations