© 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. 10-RES_POST_32K-OAI
    11. 32568
    Prev
    Next
    INDEX
    Explanations

    demonstratives (e.g., "this," "that") indicating emphasis or reference in a text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_10.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    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
    598
    +0.13
    0.6%
    41
    +0.12
    0.5%
    274
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    159
    +0.13
    0.11
    274
    +0.12
    0.09
    589
    +0.11
    0.07
    Negative Logits
    ordinary
    -1.06
    cknow
    -0.85
    undo
    -0.76
    warm
    -0.65
    unal
    -0.63
    ordinate
    -0.63
    ysical
    -0.63
    forms
    -0.61
    nor
    -0.59
    worldly
    -0.59
    POSITIVE LOGITS
     topic
    0.99
     subject
    0.84
    Got
    0.72
     came
    0.70
     question
    0.70
     relates
    0.70
     rumor
    0.69
    viously
    0.68
     saga
    0.68
     heading
    0.68
    Activations Density 0.821%

    No Known Activations