© 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. 26710
    Prev
    Next
    INDEX
    Explanations

    instances of physical aggression or violent actions

    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
    141
    +0.12
    0.5%
    373
    +0.10
    0.4%
    695
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    720
    +0.12
    0.17
    546
    +0.10
    0.14
    43
    +0.09
    0.13
    Negative Logits
     and
    -0.87
    And
    -0.82
     And
    -0.74
    and
    -0.74
     AND
    -0.72
    anded
    -0.70
    ands
    -0.64
    DonaldTrump
    -0.64
     &
    -0.64
    ortium
    -0.57
    POSITIVE LOGITS
    advertising
    0.76
     lest
    0.75
     thereby
    0.75
     whichever
    0.70
    assuming
    0.70
     respectively
    0.67
    nown
    0.67
     Alternatively
    0.64
    Alternatively
    0.62
    VERTISEMENT
    0.61
    Activations Density 1.797%

    No Known Activations