© 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. 8-RES_POST_32K-OAI
    11. 8693
    Prev
    Next
    INDEX
    Explanations

    instances of punctuation, particularly commas

    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
    259
    +0.14
    0.6%
    179
    +0.11
    0.5%
    455
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    176
    +0.14
    0.01
    284
    +0.11
    0.01
    565
    +0.10
    0.01
    Negative Logits
    hift
    -0.81
    geist
    -0.80
    hole
    -0.73
    holes
    -0.72
    hoe
    -0.69
    sectional
    -0.69
    nut
    -0.66
    ppings
    -0.65
    cipline
    -0.65
    ngth
    -0.64
    POSITIVE LOGITS
     Moroc
    0.78
     srfAttach
    0.78
     CLSID
    0.74
    ©¶æ
    0.73
     millenn
    0.69
     coron
    0.69
     Regulatory
    0.69
    soType
    0.68
    Þ
    0.65
     Reyes
    0.64
    Activations Density 0.049%

    No Known Activations