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-2-9B
    4. Attention Out - 16k
    5. 0-GEMMASCOPE-ATT-16K
    6. 1276
    Prev
    Next
    INDEX
    Explanations

    attends to the phrase "said" from various forms of expression where a speaker's assertion is involved

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-att/layer_0/width_16k/average_l0_61
    Prompts (Dashboard)
    16,384 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
    0:0.01
    1:0.01
    2:0.03
    3:0.65
    4:0.02
    5:0.03
    6:0.02
    7:0.02
    8:0.01
    9:0.01
    10:0.01
    11:0.02
    12:0.02
    13:0.03
    14:0.02
    15:0.02
    Negative Logits
     Wikiseite
    -0.42
    guiente
    -0.37
    __":
    
    -0.35
     desmotivaciones
    -0.34
     plegable
    -0.33
     Bewußt
    -0.33
     enfans
    -0.33
     idéia
    -0.32
     ciudadana
    -0.32
     Infór
    -0.32
    POSITIVE LOGITS
     S
    0.48
    ker
    0.48
    ium
    0.47
    sem
    0.46
     V
    0.46
     (
    0.46
    rod
    0.46
    cad
    0.45
    iner
    0.45
     L
    0.45
    Activations Density 0.234%

    No Known Activations