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. 113843
    Prev
    Next
    INDEX
    Explanations

    specific programming constructs and components related to database repository patterns, particularly within the context of Java

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

    programming class names

    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
     gyhoeddwyd
    -0.43
     especiais
    -0.42
    Personendaten
    -0.41
    ftagPool
    -0.39
    outWeight
    -0.39
     úteis
    -0.38
    jekk
    -0.38
     lots
    -0.38
     lengkap
    -0.38
    inSlope
    -0.37
    POSITIVE LOGITS
    yntaxException
    0.52
    ſelves
    0.51
    clyde
    0.50
     BaseActivity
    0.48
    exitRule
    0.48
    mption
    0.47
    abstract
    0.47
    ɒ
    0.46
     abstract
    0.46
     ſch
    0.45
    Activations Density 0.006%

    No Known Activations