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
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 122877
    Prev
    Next
    INDEX
    Explanations

    references to statistical data or information about organizations

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

    Technical textual elements like punctuation marks, reference notation, code syntax, and connecting words that typically appear in academic, scientific, or formal documentation.

    eleuther_acts_top20 · claude-3-7-sonnet-20250219Triggered by @lu-christina
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    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
     @"/
    -0.49
     fully
    -0.45
    gridy
    -0.42
    ↹
    -0.42
     esternos
    -0.41
     antworte
    -0.40
     Züge
    -0.40
    MigrationBuilder
    -0.40
    脚注の使い方
    -0.39
     sugli
    -0.39
    POSITIVE LOGITS
    новниш
    0.53
    XmlAccessorType
    0.48
    readystatechange
    0.45
    发表于
    0.45
    
    0.44
    новништво
    0.44
     Meksiku
    0.44
    oba
    0.42
    ніципалі
    0.42
     censiti
    0.42
    Activations Density 0.078%

    No Known Activations