© 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. 8740
    Prev
    Next
    INDEX
    Explanations

    URLs or web links

    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
    218
    +0.16
    0.7%
    314
    +0.15
    0.7%
    176
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    412
    +0.16
    0.03
    442
    +0.15
    0.02
    627
    +0.12
    0.01
    Negative Logits
     Loans
    -0.67
     intentions
    -0.67
     Scal
    -0.66
     Jarrett
    -0.66
     motivations
    -0.63
     Ethiop
    -0.61
     appearances
    -0.61
     calendars
    -0.60
     dependencies
    -0.60
     ransom
    -0.60
    POSITIVE LOGITS
    www
    1.06
    tiny
    1.02
    localhost
    0.98
    imgur
    0.98
    daily
    0.94
    bit
    0.94
    usat
    0.88
    nov
    0.87
    goo
    0.86
    steam
    0.82
    Activations Density 0.069%

    No Known Activations