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-2-2B
    4. Attention Out - 16k
    5. 0-GEMMASCOPE-ATT-16K
    6. 178
    Prev
    Next
    INDEX
    Explanations

    attends to tokens indicating visibility from tokens indicating location or condition

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 0-gemmascope-att-16k
    Configuration
    google/gemma-scope-2b-pt-att/layer_0/width_16k/average_l0_104
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.0.attn.hook_z
    Hook Layer
    0
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Head Attr Weights
    0:0.61
    1:0.02
    2:0.02
    3:0.02
    4:0.20
    5:0.03
    6:0.01
    7:0.05
    Negative Logits
     кӀ
    -0.29
    Vidite
    -0.28
    .
    -0.28
     little
    -0.25
    room
    -0.25
     maestra
    -0.25
    Allowed
    -0.25
    racite
    -0.25
     picioare
    -0.25
     lẽ
    -0.24
    POSITIVE LOGITS
     nonatomic
    0.42
     StatefulWidget
    0.41
    存于互联网档案馆
    0.40
     مشين
    0.39
    VIAF
    0.39
     LoggerFactory
    0.38
     onCreateView
    0.38
     FetchType
    0.37
     doInBackground
    0.37
    ModelSerializer
    0.36
    Activations Density 0.020%

    No Known Activations