© 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. 29584
    Prev
    Next
    INDEX
    Explanations

    mentions of authoritative figures or entities related to legal and governance contexts

    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
    99
    +0.12
    0.5%
    381
    +0.11
    0.5%
    269
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.12
    0.06
    659
    +0.11
    0.04
    99
    +0.10
    0.03
    Negative Logits
    uto
    -0.85
    amus
    -0.84
    abase
    -0.80
    ãĥŃ
    -0.79
     Giovanni
    -0.79
    andom
    -0.76
    ãĤ¤ãĥĪ
    -0.76
    acity
    -0.75
     Adin
    -0.75
    apon
    -0.75
    POSITIVE LOGITS
    e
    0.89
     Jets
    0.84
    eh
    0.78
    ez
    0.75
     Heights
    0.75
    edly
    0.74
    er
    0.73
    ey
    0.73
     Crus
    0.72
    ------------
    0.71
    Activations Density 0.409%

    No Known Activations