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. 6573
    Prev
    Next
    INDEX
    Explanations

    attends to the NMR frequency values from the isolated numeric references that follow the frequency context

    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
    0:0.11
    1:0.14
    2:0.11
    3:0.09
    4:0.13
    5:0.05
    6:0.14
    7:0.18
    Negative Logits
     People
    -0.26
     poroz
    -0.25
    oi
    -0.24
     
    -0.24
    𝘃
    -0.23
     In
    -0.23
     piscinas
    -0.22
     â
    -0.22
     kac
    -0.21
     Recruitment
    -0.21
    POSITIVE LOGITS
     оригіналу
    0.42
     متعلقه
    0.41
    !*\
    0.41
    parsedMessage
    0.41
    "":
    0.40
     ddelweddau
    0.38
    日閲覧
    0.38
    
    0.37
    tvguidetime
    0.37
    省市镇
    0.37
    Activations Density 0.117%

    No Known Activations