Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 11-GEMMASCOPE-RES-16K
    6. 13178
    Prev
    Next
    INDEX
    Explanations

    syntactic structures, particularly related to function calls and object descriptors in code

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.11.hook_resid_post
    Hook Layer
    11
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Gilli
    -0.52
    ptime
    -0.52
     กัน
    -0.51
    itag
    -0.50
    lungen
    -0.50
     hành
    -0.50
    換え
    -0.50
    woch
    -0.49
     Ot
    -0.49
    SIDE
    -0.49
    POSITIVE LOGITS
    ArgsConstructor
    0.92
    endphp
    0.90
    ImageContext
    0.88
     uParam
    0.85
     صوتيه
    0.84
    }")]
    0.80
    SharedCtor
    0.79
     useAppContext
    0.77
     propOrder
    0.75
    didSet
    0.75
    Activations Density 0.227%

    No Known Activations