© 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. 7850
    Prev
    Next
    INDEX
    Explanations

    symbols and special characters in various contexts

    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
    447
    +0.13
    0.6%
    99
    +0.11
    0.5%
    64
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    87
    +0.13
    -0.01
    584
    +0.11
    0.06
    23
    +0.11
    0.05
    Negative Logits
     categor
    -0.71
    oples
    -0.62
    boarding
    -0.61
     externally
    -0.60
     motorists
    -0.57
    wagen
    -0.57
     appointments
    -0.56
     constituents
    -0.56
     everywhere
    -0.56
     marketers
    -0.56
    POSITIVE LOGITS
     SetFontSize
    0.77
    ertodd
    0.75
     Remix
    0.74
     Learns
    0.71
     Mehran
    0.70
     Killed
    0.69
     WATCHED
    0.68
     CentOS
    0.66
     Sabha
    0.65
     ãĥ
    0.64
    Activations Density 0.441%

    No Known Activations