© 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. Residual Post - 32k
    9. 10-RES_POST_32K-OAI
    11. 29871
    Prev
    Next
    INDEX
    Explanations

    elements related to political scandals and governmental actions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_10.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    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
    713
    +0.10
    0.5%
    717
    +0.09
    0.4%
    723
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.10
    0.13
    16
    +0.09
    0.05
    616
    +0.09
    0.08
    Negative Logits
    orthy
    -0.69
    èĢħ
    -0.67
    eligible
    -0.66
    facts
    -0.66
    ABLE
    -0.64
    onyms
    -0.63
    ables
    -0.63
    wered
    -0.62
    TABLE
    -0.62
    ĭ
    -0.62
    POSITIVE LOGITS
     effort
    1.33
     crusade
    1.28
     crackdown
    1.23
     purge
    1.20
     campaign
    1.17
     takeover
    1.10
     attempt
    1.08
     blitz
    1.07
     concerted
    1.07
     bid
    1.06
    Activations Density 1.955%

    No Known Activations