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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Attention SAE Research Paper
    5. GPT2-Small
    7. Attention Out
    9. 5-ATT-KK
    11. 42646
    Prev
    Next
    INDEX
    Explanations

    instances of "walking away" or "walking out" in various contexts

    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
    2:0.07
    3:0.07
    4:0.29
    5:0.02
    6:0.07
    7:0.15
    8:0.04
    9:0.02
    10:0.09
    11:0.08
    Negative Logits
    inks
    -1.49
    cius
    -1.41
    ategories
    -1.38
     lingu
    -1.34
     TOD
    -1.32
    umber
    -1.31
    ctors
    -1.31
     informants
    -1.29
    population
    -1.29
    netflix
    -1.28
    POSITIVE LOGITS
     altogether
    2.09
     prematurely
    2.07
     undone
    1.59
     abruptly
    1.55
     entirely
    1.55
     intact
    1.50
     restored
    1.49
     completely
    1.48
     Judgment
    1.41
     overrun
    1.40
    Activations Density 0.009%

    No Known Activations