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. 24-GEMMASCOPE-RES-131K
    6. 1228
    Prev
    Next
    INDEX
    Explanations

    references to specific legal clauses or provisions

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

    legal clauses

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_24/width_131k/average_l0_97
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    getN
    -0.50
    wouldn
    -0.47
    doesn
    -0.44
    Amer
    -0.43
    Sher
    -0.43
     Watertown
    -0.42
    Kristi
    -0.42
    Histor
    -0.42
    getR
    -0.41
    didn
    -0.41
    POSITIVE LOGITS
     Clause
    1.11
     clause
    1.10
    Clause
    1.05
    clause
    1.04
     clauses
    0.94
     Clauses
    0.91
    clauses
    0.87
     Phrase
    0.65
     Clau
    0.63
     Clubhouse
    0.62
    Activations Density 0.005%

    No Known Activations