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. 18623
    Prev
    Next
    INDEX
    Explanations

    instances of violence and threats in contexts related to human rights violations

    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
    0:0.04
    1:0.09
    2:0.05
    3:0.07
    4:0.23
    5:0.06
    6:0.06
    7:0.09
    8:0.04
    9:0.04
    10:0.04
    11:0.15
    Negative Logits
    entimes
    -2.83
    yssey
    -2.56
    prisingly
    -2.50
    ̶
    -2.45
    vity
    -2.44
    ilst
    -2.39
    \":
    -2.27
    oubtedly
    -2.25
    ogether
    -2.24
    hof
    -2.23
    POSITIVE LOGITS
     Needs
    2.86
     Cannot
    2.74
     Uses
    2.74
     Subtle
    2.66
     Negative
    2.63
     Selected
    2.56
     Appears
    2.56
     Possible
    2.50
     Beware
    2.48
     Imper
    2.48
    Activations Density 0.251%

    No Known Activations