© 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. 45038
    Prev
    Next
    INDEX
    Explanations

    phrases related to ups and downs or challenges faced 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
    0:0.08
    1:0.02
    2:0.17
    3:0.04
    4:0.09
    5:0.02
    6:0.08
    7:0.25
    8:0.08
    9:0.03
    10:0.06
    11:0.04
    Negative Logits
    english
    -2.73
    Shin
    -1.82
    info
    -1.68
    licensed
    -1.63
    voice
    -1.55
    tyard
    -1.55
    INC
    -1.54
    Secure
    -1.52
    Seattle
    -1.50
     Cary
    -1.49
    POSITIVE LOGITS
     setbacks
    2.00
     factors
    1.85
     lows
    1.85
     failures
    1.79
     successes
    1.74
     pacing
    1.73
     empires
    1.66
     highs
    1.65
    scill
    1.64
     personalities
    1.63
    Activations Density 0.000%

    No Known Activations