© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 8-RES_POST_32K-OAI
    11. 9011
    Prev
    Next
    INDEX
    Explanations

    references to government departments and agencies, particularly the State Department and Pentagon

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_8.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.8.hook_resid_post
    Hook Layer
    8
    Architecture
    standard
    Context Size
    64
    Dataset
    Skylion007/openwebtext
    Activation Function
    topk
    Embeds
    IFrame
    Link
    Neuron Alignment
    Index
    Value
    % of L₁
    373
    +0.12
    0.5%
    565
    +0.10
    0.5%
    286
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    146
    +0.12
    0.04
    288
    +0.10
    0.03
    617
    +0.09
    0.02
    Negative Logits
    amaz
    -0.87
    liga
    -0.79
    ãĤ©
    -0.75
     Hallow
    -0.73
    é»Ĵ
    -0.71
    alloween
    -0.69
    AMD
    -0.68
    @#&
    -0.68
    NetMessage
    -0.66
     Measure
    -0.66
    POSITIVE LOGITS
     diplomats
    0.98
    ultz
    0.93
     envoy
    0.92
    CLASSIFIED
    0.90
     diplomatic
    0.89
     Correspond
    0.85
     embassies
    0.85
     aide
    0.85
     correspond
    0.84
     diplomat
    0.84
    Activations Density 0.117%

    No Known Activations