© 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. 2943
    Prev
    Next
    INDEX
    Explanations

    references to language learning and communication barriers

    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
    0:0.15
    1:0.02
    2:0.27
    3:0.08
    4:0.03
    5:0.09
    6:0.05
    7:0.05
    8:0.07
    9:0.06
    10:0.03
    11:0.03
    Negative Logits
     Cerberus
    -2.66
     contrace
    -2.64
     Pwr
    -2.60
     RAID
    -2.60
     Rove
    -2.58
     insured
    -2.57
     mattress
    -2.55
     GEAR
    -2.55
    destroy
    -2.54
     Vapor
    -2.53
    POSITIVE LOGITS
     bilingual
    5.69
     lingu
    5.62
     linguistic
    5.37
    Language
    5.26
     translator
    5.23
    ilingual
    5.22
     pronunciation
    5.09
     languages
    5.07
     Language
    5.05
     grammar
    4.95
    Activations Density 0.453%

    No Known Activations