© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 9018
    Prev
    Next
    INDEX
    Explanations

    direct quotations or attributions in textual content

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Neuron Alignment
    Index
    Value
    % of L₁
    295
    +0.12
    0.6%
    426
    +0.10
    0.5%
    27
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    426
    +0.12
    0.09
    734
    +0.10
    0.07
    352
    +0.09
    0.07
    Negative Logits
    ãĥij
    -0.80
     FORM
    -0.70
     stret
    -0.70
     Playstation
    -0.69
    DVD
    -0.68
     SUPER
    -0.67
    ãĥ¯ãĥ³
    -0.66
    clamation
    -0.66
    MEN
    -0.64
     deed
    -0.64
    POSITIVE LOGITS
     concedes
    1.11
     recalls
    0.99
     says
    0.98
     observes
    0.93
     admits
    0.93
     argues
    0.92
     Mehran
    0.90
     explains
    0.88
     Mayer
    0.83
     Fein
    0.82
    Activations Density 0.494%

    No Known Activations