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-2-2B
    4. Attention Out - 16k
    5. 0-GEMMASCOPE-ATT-16K
    6. 122
    Prev
    Next
    INDEX
    Explanations

    attends to the token "revealed" from several different contexts or claims made earlier 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
    0:0.51
    1:0.01
    2:0.01
    3:0.01
    4:0.31
    5:0.04
    6:0.01
    7:0.06
    Negative Logits
    ifoli
    -0.37
    MigrationBuilder
    -0.36
    <?
    -0.36
     Extragalactic
    -0.35
    dehydes
    -0.35
    ingale
    -0.35
     ourselves
    -0.31
    исленность
    -0.31
    bernate
    -0.31
    æus
    -0.31
    POSITIVE LOGITS
     doInBackground
    0.37
    AutoField
    0.33
    Jegyzetek
    0.32
     resourceCulture
    0.30
    subplots
    0.30
    Literatur
    0.30
    zeera
    0.29
    $)
    0.29
    Weblinks
    0.29
    Einzelnachweise
    0.29
    Activations Density 0.042%

    No Known Activations