Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Under Peer Review · Attention SAE Research Paper
    3. GPT2-Small
    4. Attention Out
    5. 9-ATT-KK
    6. 11448
    Prev
    Next
    INDEX
    Explanations

    statements congratulating accomplishments or records

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-att-kk
    Configuration
    ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L9_Hcat_z_lr1.20e-03_l11.20e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    float32
    Hook Name
    blocks.9.attn.hook_z
    Hook Layer
    9
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Head Attr Weights
    0:0.05
    1:0.03
    2:0.04
    3:0.04
    4:0.08
    5:0.04
    6:0.23
    7:0.06
    8:0.11
    9:0.19
    10:0.03
    11:0.04
    Negative Logits
     Reynolds
    -3.75
    Todd
    -3.46
     Pilot
    -3.26
     Wing
    -3.13
     cess
    -3.13
     VID
    -3.10
     Benson
    -3.09
     Twin
    -3.07
    hover
    -3.07
    NZ
    -3.06
    POSITIVE LOGITS
     Barcelona
    8.17
     Messi
    7.77
     Catalan
    7.04
     Catalonia
    6.79
     Guardiola
    6.71
    celona
    6.43
     Catal
    6.28
    Catal
    6.00
     Lionel
    5.58
    ça
    5.45
    Activations Density 0.008%

    No Known Activations