© 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. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 8-RES_POST_32K-OAI
    11. 6586
    Prev
    Next
    INDEX
    Explanations

    phrases and terms that indicate relationships or connections among people and groups

    oai_token-act-pair · gpt-4o-miniTriggered 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
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    476
    +0.11
    0.5%
    720
    +0.10
    0.5%
    676
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    393
    +0.11
    0.10
    363
    +0.10
    0.07
    720
    +0.08
    0.09
    Negative Logits
    ï¸
    -0.88
    mentioned
    -0.77
    hog
    -0.68
     WRITE
    -0.65
    wright
    -0.65
    BUR
    -0.62
     Turing
    -0.62
     Canaver
    -0.62
    vable
    -0.61
    usive
    -0.61
    POSITIVE LOGITS
     federation
    0.92
     kins
    0.91
     alliances
    0.88
     Individual
    0.86
     individuals
    0.86
     ethnicity
    0.86
     territorial
    0.86
    community
    0.85
     individual
    0.83
     solidarity
    0.83
    Activations Density 1.233%

    No Known Activations