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

    Neuronpedia

    Jacobian LensNEW
    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. 41773
    Prev
    Next
    INDEX
    Explanations

    conjunctions, particularly variations of the word "and."

    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
    0:0.02
    1:0.02
    2:0.11
    3:0.04
    4:0.08
    5:0.05
    6:0.21
    7:0.06
    8:0.09
    9:0.05
    10:0.12
    11:0.09
    Negative Logits
     hindsight
    -1.53
    hani
    -1.49
     nutshell
    -1.35
    ��
    -1.29
     Sauce
    -1.28
    QL
    -1.26
    BLIC
    -1.25
     Tome
    -1.23
     Redux
    -1.20
    Adds
    -1.17
    POSITIVE LOGITS
    artifacts
    1.57
    izont
    1.49
    wcs
    1.47
    accompanied
    1.46
    thinkable
    1.42
    interrupted
    1.40
    Stage
    1.32
    cycle
    1.30
     spor
    1.29
    enough
    1.27
    Activations Density 0.040%

    No Known Activations