© 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. 6683
    Prev
    Next
    INDEX
    Explanations

    phrases related to moderation, banning, and user guidelines in online platforms

    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
    218
    +0.15
    0.7%
    752
    +0.11
    0.5%
    35
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    570
    +0.15
    0.00
    298
    +0.11
    0.01
    211
    +0.10
    0.00
    Negative Logits
    iban
    -0.76
     analogue
    -0.67
    uxe
    -0.64
    lation
    -0.64
    lig
    -0.63
    ulton
    -0.63
     substitution
    -0.62
     Canaver
    -0.62
    omics
    -0.62
     privatization
    -0.62
    POSITIVE LOGITS
    cheat
    0.70
    GU
    0.65
     Hide
    0.64
     rad
    0.58
    conduct
    0.58
     shields
    0.57
    chy
    0.57
    RF
    0.56
     fuzz
    0.56
    arget
    0.55
    Activations Density 0.130%

    No Known Activations