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. 9-ATT-KK
    6. 10804
    Prev
    Next
    INDEX
    Explanations

    phrases that highlight affiliations and positions of individuals or entities in a political context

    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
    0:0.10
    1:0.05
    2:0.11
    3:0.12
    4:0.04
    5:0.06
    6:0.03
    7:0.27
    8:0.05
    9:0.03
    10:0.06
    11:0.03
    Negative Logits
    Advertisements
    -3.09
    blocks
    -2.74
    amins
    -2.52
    >]
    -2.52
    nas
    -2.51
    Direct
    -2.49
    】
    -2.33
    Blocks
    -2.33
     =====
    -2.33
     blocks
    -2.29
    POSITIVE LOGITS
     successor
    3.39
     revived
    3.35
     whistleblower
    3.06
     resurrected
    2.92
     reformed
    2.84
     rebuilt
    2.83
     retired
    2.75
     rehabilit
    2.72
     lobbyist
    2.65
     reunited
    2.58
    Activations Density 0.011%

    No Known Activations