© 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. 7467
    Prev
    Next
    INDEX
    Explanations

    punctuation, specifically the comma character

    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
    102
    +0.20
    0.9%
    55
    +0.14
    0.7%
    7
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    102
    +0.20
    0.05
    55
    +0.14
    0.04
    7
    +0.09
    0.03
    Negative Logits
    ortmund
    -0.77
     Telescope
    -0.74
    regate
    -0.72
    Names
    -0.72
     Fighters
    -0.70
    inav
    -0.70
     deductions
    -0.70
    ESE
    -0.69
    Previous
    -0.69
    çīĪ
    -0.68
    POSITIVE LOGITS
     indistinguishable
    0.99
     irrelevant
    0.94
     synonymous
    0.93
     incapable
    0.92
     insepar
    0.92
     supposed
    0.90
     impossible
    0.90
     extremely
    0.89
     riddled
    0.89
     barred
    0.87
    Activations Density 0.190%

    No Known Activations