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. 46117
    Prev
    Next
    INDEX
    Explanations

    code functionalities related to reading, writing, and manipulating data structures

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

    uniform4fv data types

    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
    })));
    -0.32
    WithValue
    -0.31
    ())->
    -0.31
     leeren
    -0.30
    ')->
    -0.30
     Mu
    -0.30
    ══
    -0.30
    '):
    
    -0.29
    "/>
    
    -0.29
     "");
    
    -0.29
    POSITIVE LOGITS
    rawDesc
    0.73
    يكب
    0.66
    rungsseite
    0.66
    NameInMap
    0.64
     CallOverrides
    0.63
    Бахар
    0.59
     ujednoznacz
    0.58
     AssemblyCulture
    0.58
     ſch
    0.57
    glGen
    0.57
    Activations Density 0.017%

    No Known Activations