© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Attention SAE Research Paper
    5. GPT2-Small
    7. Attention Out
    9. 9-ATT-KK
    11. 10523
    Prev
    Next
    INDEX
    Explanations

    statements related to reports and findings, particularly those discussing oversight, safety, and regulations

    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
    0:0.05
    1:0.03
    2:0.11
    3:0.17
    4:0.04
    5:0.33
    6:0.02
    7:0.06
    8:0.05
    9:0.02
    10:0.06
    11:0.01
    Negative Logits
     astronomer
    -2.60
     Ragnar
    -2.51
     Dalai
    -2.50
    blogspot
    -2.47
     Investigator
    -2.40
     Einstein
    -2.39
     Nobel
    -2.36
     commentary
    -2.36
     Inquisitor
    -2.34
     Pulitzer
    -2.31
    POSITIVE LOGITS
     deteriorated
    2.92
     pervasive
    2.83
     inefficient
    2.82
     worsened
    2.68
     reliance
    2.64
     inconsist
    2.64
     decreased
    2.63
     impaired
    2.62
     Lack
    2.61
     deteriorating
    2.61
    Activations Density 0.329%

    No Known Activations