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
    4. Residual Stream - 16k
    5. 12-GEMMASCOPE-RES-16K
    6. 2832
    Prev
    Next
    INDEX
    Explanations

    occurrences of string literals and data structures in code, particularly those related to object properties

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 12-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_12/width_16k/average_l0_130
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.12.hook_resid_post
    Hook Layer
    12
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     resourceCulture
    -0.56
    parsedMessage
    -0.50
    WithMany
    -0.49
    BagConstraints
    -0.48
    tagHelperRunner
    -0.48
    
    -0.47
    
    -0.46
    UIControlState
    -0.46
     inalámbrica
    -0.43
    makeText
    -0.43
    POSITIVE LOGITS
    verwijspagina
    0.39
    beans
    0.38
    mtrl
    0.37
    noscript
    0.36
     Italijanski
    0.36
    hasOwnProperty
    0.36
    ActionCreators
    0.36
    dsquared
    0.35
     bon
    0.35
    ovsky
    0.35
    Activations Density 0.423%

    No Known Activations