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. 0-GEMMASCOPE-ATT-16K
    6. 6114
    Prev
    Next
    INDEX
    Explanations

    attends to the expression of personal opinions or thoughts, marked by instances of the token "I," from various subjects and contexts later in the text

    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
    0:0.03
    1:0.02
    2:0.03
    3:0.01
    4:0.09
    5:0.76
    6:0.02
    7:0.02
    Negative Logits
     don
    -0.66
     am
    -0.57
     her
    -0.52
     didn
    -0.51
     can
    -0.49
     just
    -0.47
     who
    -0.47
     she
    -0.46
     have
    -0.45
     had
    -0.44
    POSITIVE LOGITS
    GEBURTSDATUM
    0.82
    出版年
    0.60
    JspWriter
    0.60
    afficheront
    0.56
    TestingModule
    0.54
    httphttps
    0.54
    EDEFAULT
    0.53
    sizeCache
    0.52
     PyLong
    0.50
    elemField
    0.50
    Activations Density 1.582%

    No Known Activations