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-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 85825
    Prev
    Next
    INDEX
    Explanations

    comments and documentation structures in code

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

    compiled code comments

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    tagHelperRunner
    -0.80
     nakalista
    -0.75
    StoryboardSegue
    -0.75
    ⓧ
    -0.69
    RenderAtEndOf
    -0.65
    AISSEE
    -0.65
     насељу
    -0.62
    webElementXpaths
    -0.59
    TagMode
    -0.58
    RegressionTest
    -0.58
    POSITIVE LOGITS
     *
    0.48
     **
    0.38
     /
    0.36
     noted
    0.34
    **
    0.33
    labelledby
    0.33
     weight
    0.33
    ///**
    0.32
    tées
    0.32
     *
    
    0.31
    Activations Density 0.032%

    No Known Activations