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. 280
    Prev
    Next
    INDEX
    Explanations

    attends to the token "println" from tokens that begin with a quotation mark

    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
    0:0.65
    1:0.02
    2:0.02
    3:0.02
    4:0.12
    5:0.09
    6:0.01
    7:0.03
    Negative Logits
    (
    -0.31
    SequentialGroup
    -0.31
    our
    -0.31
    _
    -0.28
    but
    -0.27
    ((
    -0.27
     estup
    -0.27
     costitu
    -0.26
     layui
    -0.26
    ()
    -0.26
    POSITIVE LOGITS
     Wikimedijinoj
    0.37
    لينكات
    0.35
    ])[
    0.35
     disponibilités
    0.35
     CreateTagHelper
    0.35
    anager
    0.35
    msgTypes
    0.34
     للاسماء
    0.34
    dafx
    0.34
    lgari
    0.33
    Activations Density 0.009%

    No Known Activations