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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Attention Out - 32k
    9. 3-ATT_32K-OAI
    11. 4469
    Prev
    Next
    INDEX
    Explanations

    references to specific individuals and cultural contexts, particularly related to East Asia

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-att_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-attn-out-SAEs/v5_32k_layer_3
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.3.hook_attn_out
    Hook Layer
    3
    Architecture
    standard
    Context Size
    64
    Dataset
    Skylion007/openwebtext
    Activation Function
    topk
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    343
    +0.14
    0.6%
    154
    +0.11
    0.5%
    149
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    343
    +0.14
    0.03
    521
    +0.11
    0.02
    284
    +0.09
    0.01
    Negative Logits
     flares
    -0.80
     flare
    -0.74
    eele
    -0.68
     slopes
    -0.67
    ensor
    -0.66
    agle
    -0.65
    ngth
    -0.63
    brance
    -0.62
     injection
    -0.62
     Sloan
    -0.60
    POSITIVE LOGITS
    merce
    0.98
     Reincarn
    0.87
    Wan
    0.74
    urst
    0.73
    EStreamFrame
    0.71
    hua
    0.71
     Juliet
    0.71
    atar
    0.70
    acea
    0.69
     Ghana
    0.68
    Activations Density 0.134%

    No Known Activations