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-2-2B
    4. Attention Out - 16k
    5. 10-GEMMASCOPE-ATT-16K
    6. 6862
    Prev
    Next
    INDEX
    Explanations

    attends to the token "must" from related tokens referring to actions or recommendations

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 10-gemmascope-att-16k
    Configuration
    google/gemma-scope-2b-pt-att/layer_10/width_16k/average_l0_70
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.10.attn.hook_z
    Hook Layer
    10
    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
    0:0.15
    1:0.20
    2:0.14
    3:0.08
    4:0.04
    5:0.02
    6:0.05
    7:0.28
    Negative Logits
    tawesome
    -0.34
    MLLoader
    -0.34
    Хьажоргаш
    -0.30
    providedIn
    -0.29
    Referencie
    -0.29
     jegy
    -0.28
    UnsafeEnabled
    -0.28
     ...
    
    -0.27
    hésite
    -0.27
     brancas
    -0.27
    POSITIVE LOGITS
     tủ
    0.27
     tartalomajánló
    0.26
    ftagPool
    0.25
    oraf
    0.25
     Fels
    0.23
    anke
    0.22
    tshell
    0.22
    }}}$
    0.22
    orius
    0.22
    esthetics
    0.21
    Activations Density 0.055%

    No Known Activations