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. Under Peer Review · Attention SAE Research Paper
    3. GPT2-Small
    4. Attention Out
    5. 5-ATT-KK
    6. 19400
    Prev
    Next
    INDEX
    Explanations

    punctuation marks, specifically periods and exclamation points

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 5-att-kk
    Configuration
    ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L5_Hcat_z_lr1.20e-03_l11.00e+00_ds49152_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    Data Type
    float32
    Hook Name
    blocks.5.attn.hook_z
    Hook Layer
    5
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Head Attr Weights
    0:0.10
    1:0.42
    2:0.03
    3:0.03
    4:0.02
    5:0.07
    6:0.02
    7:0.01
    8:0.09
    9:0.09
    10:0.03
    11:0.04
    Negative Logits
    asser
    -1.60
    uced
    -1.59
     Rankings
    -1.57
    esar
    -1.51
    esian
    -1.48
    ue
    -1.48
    ased
    -1.46
    engu
    -1.46
     outp
    -1.46
     Tonight
    -1.44
    POSITIVE LOGITS
    },{"
    1.95
    conservancy
    1.76
    ↵
    1.64
    �
    1.61
    SPONSORED
    1.61
    Footnote
    1.56
     CrossRef
    1.54
    });
    1.54
    ohm
    1.46
     Wick
    1.46
    Activations Density 0.079%

    No Known Activations