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. 9-ATT-KK
    6. 11413
    Prev
    Next
    INDEX
    Explanations

    concepts related to physics and energy transfer phenomena

    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
    0:0.04
    1:0.06
    2:0.12
    3:0.07
    4:0.19
    5:0.05
    6:0.06
    7:0.14
    8:0.03
    9:0.05
    10:0.07
    11:0.06
    Negative Logits
    retty
    -2.21
    aido
    -2.19
     TBA
    -2.14
    ahan
    -2.11
    ificate
    -2.10
     Mub
    -2.08
    intend
    -2.06
    arie
    -2.04
     boarded
    -2.02
    WB
    -2.01
    POSITIVE LOGITS
    Researchers
    3.38
    Once
    3.08
    Scientists
    3.04
    Eventually
    3.03
    Each
    3.02
    Thus
    3.00
    Engineers
    2.90
    Depending
    2.89
    Alternatively
    2.88
    Because
    2.86
    Activations Density 0.309%

    No Known Activations