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. 6540
    Prev
    Next
    INDEX
    Explanations

    attends to empty tokens from arbitrary later tokens

    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
    0:0.36
    1:0.31
    2:0.07
    3:0.04
    4:0.04
    5:0.03
    6:0.03
    7:0.06
    Negative Logits
     <<<<<<<<<<<<<<
    -0.35
     mẽ
    -0.35
    期刊论文
    -0.33
    cinoma
    -0.32
    msgSender
    -0.31
    venirs
    -0.30
    MergeFrom
    -0.30
    Hauptartikel
    -0.30
    pausal
    -0.29
    acum
    -0.28
    POSITIVE LOGITS
    )))));
    0.43
    ."));
    0.42
    ()]);
    0.41
    "]}
    0.36
    !");
    0.36
    fjspx
    0.36
    --]
    0.35
     betweenstory
    0.35
    ."),
    0.35
     {}));
    0.35
    Activations Density 1.487%

    No Known Activations