© 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 EvalsExportsAPI 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. 28528
    Prev
    Next
    INDEX
    Explanations

    references to images and visual representations

    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
    66
    +0.14
    0.6%
    218
    +0.10
    0.5%
    687
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    66
    +0.14
    0.12
    764
    +0.10
    0.05
    640
    +0.10
    0.14
    Negative Logits
    ternity
    -0.73
    alloc
    -0.65
     abstinence
    -0.63
     Reincarn
    -0.62
    acs
    -0.61
    doms
    -0.61
    ebook
    -0.60
    exempt
    -0.60
    opoly
    -0.59
    mith
    -0.59
    POSITIVE LOGITS
     photo
    1.88
     picture
    1.79
     photograph
    1.76
     image
    1.70
     photos
    1.67
     pictures
    1.61
    picture
    1.61
     screenshot
    1.60
     pics
    1.60
     Photo
    1.55
    Activations Density 4.699%

    No Known Activations