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. 3038
    Prev
    Next
    INDEX
    Explanations

    sentences that convey negative news or reports of unfortunate events

    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
    0:0.02
    1:0.07
    2:0.03
    3:0.04
    4:0.09
    5:0.08
    6:0.10
    7:0.08
    8:0.05
    9:0.03
    10:0.03
    11:0.30
    Negative Logits
    ♦
    -3.70
    •
    -3.55
    …………
    -3.48
    .—
    -3.44
    ―
    -3.41
    ●
    -3.40
    ————————————————
    -3.40
    «
    -3.33
    )</
    -3.33
    CLASSIFIED
    -3.32
    POSITIVE LOGITS
     Mobil
    2.27
     Seas
    2.25
     Atomic
    2.23
     Gol
    2.21
     Airl
    2.18
    ↵
    2.15
     Defence
    2.12
     Nug
    2.11
     Str
    2.07
     Starter
    2.05
    Activations Density 0.009%

    No Known Activations