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. 20-GEMMASCOPE-RES-16K
    6. 15668
    Prev
    Next
    INDEX
    Explanations

    mentions of interactive elements or processes within contexts

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

    In-person interviews or in-situ treatments

    np_acts-logits-general · gemini-2.0-flash

    in- followed by context

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     onCancelled
    -0.67
    脚注の使い方
    -0.67
    numerusform
    -0.65
    astéroïdes
    -0.60
    bildeten
    -0.59
    shadowOpacity
    -0.59
    Становништво
    -0.58
    tanleria
    -0.58
     Италијани
    -0.56
    تقاوى
    -0.56
    POSITIVE LOGITS
    fromnode
    0.56
    -
    0.55
    person
    0.51
     Normdatei
    0.50
    AddTagHelper
    0.49
     person
    0.47
     situ
    0.46
     vivo
    0.46
    IActionResult
    0.44
     season
    0.43
    Activations Density 0.132%

    No Known Activations