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. 5-ATT-KK
    6. 3056
    Prev
    Next
    INDEX
    Explanations

    expressions of sadness and regret

    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
    0:0.03
    1:0.02
    2:0.29
    3:0.12
    4:0.10
    5:0.03
    6:0.04
    7:0.12
    8:0.04
    9:0.03
    10:0.06
    11:0.07
    Negative Logits
     skills
    -1.91
    ograms
    -1.85
     likes
    -1.79
    ogram
    -1.62
     flair
    -1.62
    ヘ
    -1.60
    rities
    -1.55
     preference
    -1.53
    �
    -1.50
     IQ
    -1.49
    POSITIVE LOGITS
    ilst
    1.81
     Stra
    1.66
    abis
    1.52
    haus
    1.52
     SECTION
    1.49
    olid
    1.48
    dy
    1.43
    onel
    1.43
    theless
    1.38
    held
    1.37
    Activations Density 0.001%

    No Known Activations