© 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 EvalsExportsinterp-engineNEWAPI 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. 32521
    Prev
    Next
    INDEX
    Explanations

    references to network security vulnerabilities and attacks

    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
    660
    +0.13
    0.6%
    251
    +0.10
    0.5%
    574
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.13
    0.09
    718
    +0.10
    0.08
    418
    +0.10
    0.04
    Negative Logits
    Safety
    -0.74
     Goddess
    -0.68
     Excellence
    -0.67
     Calm
    -0.67
    Compl
    -0.64
     Volvo
    -0.64
     condol
    -0.63
    Case
    -0.63
    HUD
    -0.62
    illin
    -0.62
    POSITIVE LOGITS
     accessed
    1.08
     imperson
    1.04
     spoof
    1.01
     decrypt
    0.96
     triv
    0.95
     steal
    0.91
     malicious
    0.90
     unauthorized
    0.89
     hijacked
    0.89
     remotely
    0.89
    Activations Density 1.375%

    No Known Activations