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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 105156
    Prev
    Next
    INDEX
    Explanations

    comparative expressions and descriptions that highlight differences in quality or characteristics

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
     aDecoder
    -0.55
    LabelTagHelper
    -0.54
     propOrder
    -0.54
    pushFollow
    -0.48
     invokingState
    -0.48
    المشاركات
    -0.48
    Спољашње
    -0.48
    jsxFileName
    -0.47
    inheritDoc
    -0.46
     endwhile
    -0.46
    POSITIVE LOGITS
     eikä
    0.36
    AddTagHelper
    0.35
     inkább
    0.35
     TextInputType
    0.35
     selera
    0.34
     appunto
    0.34
    ParallelGroup
    0.34
     fun
    0.34
    适合
    0.34
     définitivement
    0.33
    Activations Density 0.034%

    No Known Activations