Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Under Peer Review · Attention SAE Research Paper
    3. GPT2-Small
    4. Attention Out
    5. 3-ATT-KK
    6. 10065
    Prev
    Next
    INDEX
    Explanations

    subjects and pronouns in relation to actions and states

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-att-kk
    Configuration
    ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L3_Hcat_z_lr1.20e-03_l19.00e-01_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.3.attn.hook_z
    Hook Layer
    3
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Head Attr Weights
    0:0.06
    1:0.14
    2:0.13
    3:0.05
    4:0.07
    5:0.06
    6:0.10
    7:0.10
    8:0.07
    9:0.04
    10:0.07
    11:0.06
    Negative Logits
    etheless
    -1.47
     proble
    -1.28
    eatures
    -1.21
    bda
    -1.17
    eston
    -1.16
    furt
    -1.13
    acas
    -1.12
     misunder
    -1.10
    assies
    -1.09
     agre
    -1.09
    POSITIVE LOGITS
    \":
    1.30
    アル
    1.21
    icide
    1.14
    ー
    1.10
    icides
    1.07
    −
    1.05
    ウ
    1.00
     �
    0.99
     decade
    0.97
     Trans
    0.97
    Activations Density 0.129%

    No Known Activations