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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Under Peer Review · Attention SAE Research Paper
    3. GPT2-Small
    4. Attention Out
    5. 2-ATT-KK
    6. 22427
    Prev
    Next
    INDEX
    Explanations

    legal and criminal terms and concepts

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot

    legal and criminal challenges facing individuals or groups

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-att-kk
    Configuration
    ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L2_Hcat_z_lr1.20e-03_l11.00e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v4.pt
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    float32
    Hook Name
    blocks.2.attn.hook_z
    Hook Layer
    2
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Head Attr Weights
    0:0.07
    1:0.03
    2:0.05
    3:0.06
    4:0.04
    5:0.08
    6:0.03
    7:0.03
    8:0.43
    9:0.05
    10:0.06
    11:0.02
    Negative Logits
     contrace
    -1.74
    ufact
    -1.73
    asonic
    -1.64
    umin
    -1.48
    ovies
    -1.45
     sexes
    -1.39
    glomer
    -1.37
     slides
    -1.31
     interacts
    -1.27
     shalt
    -1.26
    POSITIVE LOGITS
     custod
    1.46
     loophole
    1.45
     negligence
    1.41
     hurdle
    1.39
     hurdles
    1.36
    gerald
    1.34
     difficulties
    1.31
     shortfall
    1.30
     bankruptcy
    1.30
     dilemma
    1.28
    Activations Density 0.034%

    No Known Activations