© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    3. GPT2-Small
    4. Residual Post - 32k
    5. 10-RES_POST_32K-OAI
    6. 29930
    Prev
    Next
    INDEX
    Explanations

    numerical values and statistics related to costs or expenses

    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
    373
    +0.16
    0.7%
    742
    +0.10
    0.4%
    464
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    539
    +0.16
    0.18
    640
    +0.10
    0.16
    143
    +0.09
    0.15
    Negative Logits
    Newsletter
    -0.73
    bryce
    -0.66
    bean
    -0.65
    qu
    -0.59
    interstitial
    -0.59
    hner
    -0.58
    tml
    -0.56
     sshd
    -0.56
    hig
    -0.56
    SPONSORED
    -0.55
    POSITIVE LOGITS
    ĸļ
    0.86
    Í
    0.67
    %]
    0.67
     !!
    0.67
    JUST
    0.64
     ï
    0.62
    orious
    0.62
    %"
    0.62
    urated
    0.61
    iless
    0.61
    Activations Density 1.164%

    No Known Activations