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. 6136
    Prev
    Next
    INDEX
    Explanations

    attends to numeric expressions or values from mathematical or scientific contexts, particularly focusing on those that are squared or raised to powers

    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
    0:0.27
    1:0.05
    2:0.03
    3:0.03
    4:0.15
    5:0.03
    6:0.09
    7:0.31
    Negative Logits
    ↵↵
    -0.37
    .
    -0.35
    <strong>
    -0.35
    p
    -0.31
    and
    -0.30
    the
    -0.30
    2
    -0.30
    ↵
    -0.30
    r
    -0.29
    h
    -0.29
    POSITIVE LOGITS
     CreateTagHelper
    0.48
    drawal
    0.47
     myſelf
    0.47
    IBLIO
    0.46
     كومونز
    0.46
    VERTISEMENT
    0.46
    VIAF
    0.45
    vantages
    0.45
     Normdatei
    0.44
    CHREIB
    0.44
    Activations Density 0.026%

    No Known Activations