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. 44193
    Prev
    Next
    INDEX
    Explanations

    numerous references to loss and elections

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    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
    0:0.01
    1:0.01
    2:0.05
    3:0.05
    4:0.07
    5:0.02
    6:0.03
    7:0.50
    8:0.03
    9:0.02
    10:0.08
    11:0.07
    Negative Logits
    Statement
    -1.69
    Minimum
    -1.65
    NESS
    -1.57
    Commerce
    -1.56
    Website
    -1.52
    iated
    -1.49
    Enabled
    -1.45
    Required
    -1.45
    Press
    -1.44
    女
    -1.42
    POSITIVE LOGITS
     Loss
    1.76
     Kand
    1.65
     Decay
    1.52
     landsl
    1.51
     Hurricanes
    1.49
     Kingston
    1.48
     Lost
    1.45
     Lank
    1.44
     Cantor
    1.42
     Fatal
    1.40
    Activations Density 0.050%

    No Known Activations