© 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. 29863
    Prev
    Next
    INDEX
    Explanations

    references to sin or sinful behavior

    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
    130
    +0.10
    0.5%
    288
    +0.09
    0.4%
    659
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.10
    0.02
    314
    +0.09
    -0.00
    288
    +0.09
    0.02
    Negative Logits
    theless
    -0.70
    yright
    -0.67
    è¦ļéĨĴ
    -0.63
    tin
    -0.63
    tsky
    -0.60
     females
    -0.60
     teens
    -0.60
     autism
    -0.58
    âĨij
    -0.57
    ries
    -0.57
    POSITIVE LOGITS
    atra
    1.43
    ister
    1.26
    estro
    1.13
    clair
    1.12
    ners
    0.97
    auer
    0.95
    bad
    0.93
    omon
    0.91
    ohyd
    0.90
    isters
    0.89
    Activations Density 0.140%

    No Known Activations