© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 33-GEMMASCOPE-RES-16K
    6. 11037
    Prev
    Next
    INDEX
    Explanations

    references to datasets and computational performance metrics

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 33-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_33/width_16k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.33.hook_resid_post
    Hook Layer
    33
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     ब्रेकडाउन
    -0.38
    pyx
    -0.30
    発
    -0.30
     cận
    -0.28
     score
    -0.28
    
    -0.27
     olha
    -0.27
    łecz
    -0.26
    $('.
    -0.24
    umlu
    -0.24
    POSITIVE LOGITS
     dataset
    1.04
     datasets
    0.97
     Dataset
    0.92
    数据集
    0.90
    Dataset
    0.82
     Datasets
    0.81
     benchmark
    0.79
    dataset
    0.79
    datasets
    0.77
    DATASET
    0.75
    Activations Density 0.884%

    No Known Activations