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. 7753
    Prev
    Next
    INDEX
    Explanations

    concepts related to psychological and emotional states or conditions

    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
    0:0.03
    1:0.05
    2:0.07
    3:0.03
    4:0.02
    5:0.12
    6:0.09
    7:0.13
    8:0.08
    9:0.05
    10:0.14
    11:0.13
    Negative Logits
    ギ
    -1.21
    ּ
    -1.21
    dit
    -1.14
     Olympia
    -1.10
    keleton
    -1.08
     Adin
    -1.07
    DERR
    -1.07
    thia
    -1.06
     convict
    -1.04
    �
    -1.03
    POSITIVE LOGITS
     respectively
    1.90
    oven
    1.21
    depending
    1.21
     respective
    1.15
    Characters
    1.13
    reviewed
    1.13
    These
    1.11
    ergy
    1.09
    ependent
    1.08
     collectively
    1.07
    Activations Density 0.343%

    No Known Activations